diff --git a/.github/workflows/pr-test-xeon.yml b/.github/workflows/pr-test-xeon.yml
index 683e98e47..cc9a3940e 100644
--- a/.github/workflows/pr-test-xeon.yml
+++ b/.github/workflows/pr-test-xeon.yml
@@ -90,7 +90,7 @@ jobs:
           # full machine, no partitioning. 4 files: 3 TP + autoround.
           - runner: xeon-gnr
             role: gnr
-            suite: "base-b-tp-test-cpu"
+            suite: "stage-a-tp-test-cpu-intel"
             name: gnr
 
           # Three SPR boxes each run ONE full-machine container taking one
@@ -102,19 +102,19 @@ jobs:
           # per box keeps each server within the box's memory budget.
           - runner: xeon-spr
             role: spr
-            suite: "base-b-test-cpu,base-c-test-cpu"
+            suite: "stage-a-test-cpu-intel,stage-b-test-cpu-intel"
             name: spr1
             part_id: 0
             part_size: 3
           - runner: xeon-spr
             role: spr
-            suite: "base-b-test-cpu,base-c-test-cpu"
+            suite: "stage-a-test-cpu-intel,stage-b-test-cpu-intel"
             name: spr2
             part_id: 1
             part_size: 3
           - runner: xeon-spr
             role: spr
-            suite: "base-b-test-cpu,base-c-test-cpu"
+            suite: "stage-a-test-cpu-intel,stage-b-test-cpu-intel"
             name: spr3
             part_id: 2
             part_size: 3
diff --git a/test/registered/bench_fn/test_benchmark_datasets_api.py b/test/registered/bench_fn/test_benchmark_datasets_api.py
index 9cc5d23b4..2fc659aaf 100644
--- a/test/registered/bench_fn/test_benchmark_datasets_api.py
+++ b/test/registered/bench_fn/test_benchmark_datasets_api.py
@@ -61,7 +61,7 @@ from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.test_utils import CustomTestCase
 
 register_cpu_ci(est_time=30, suite="base-a-test-cpu")
-register_cpu_ci(est_time=46, suite="base-c-test-cpu")
+register_cpu_ci(est_time=46, suite="stage-b-test-cpu-intel")
 
 
 _BENCH_SERVING_CLI_CASES = {
diff --git a/test/registered/cpu/quant/test_autoround.py b/test/registered/cpu/quant/test_autoround.py
index 767196c3d..7e954af71 100644
--- a/test/registered/cpu/quant/test_autoround.py
+++ b/test/registered/cpu/quant/test_autoround.py
@@ -24,7 +24,7 @@ from sglang.test.test_utils import (
     popen_launch_server,
 )
 
-register_cpu_ci(est_time=183, suite="base-b-tp-test-cpu")
+register_cpu_ci(est_time=183, suite="stage-a-tp-test-cpu-intel")
 
 
 class TestAutoRoundCPUConfig(CustomTestCase):
diff --git a/test/registered/cpu/test_activation.py b/test/registered/cpu/test_activation.py
index 7b6cc893c..c2beb2760 100644
--- a/test/registered/cpu/test_activation.py
+++ b/test/registered/cpu/test_activation.py
@@ -8,7 +8,7 @@ from sglang.srt.server_args import ServerArgs, set_global_server_args_for_schedu
 from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.cpu_test_utils import GeluAndMul, SiluAndMul, precision
 
-register_cpu_ci(est_time=9, suite="base-b-test-cpu")
+register_cpu_ci(est_time=9, suite="stage-a-test-cpu-intel")
 register_cpu_ci(est_time=10, suite="base-b-test-cpu-arm64")
 
 torch.manual_seed(1234)
diff --git a/test/registered/cpu/test_binding.py b/test/registered/cpu/test_binding.py
index 26e529a37..045829730 100644
--- a/test/registered/cpu/test_binding.py
+++ b/test/registered/cpu/test_binding.py
@@ -11,7 +11,7 @@ from sglang.srt.utils.numa_utils import init_threads_binding
 from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.test_utils import CustomTestCase
 
-register_cpu_ci(est_time=6, suite="base-b-tp-test-cpu")
+register_cpu_ci(est_time=6, suite="stage-a-tp-test-cpu-intel")
 
 
 class TestBinding(CustomTestCase):
diff --git a/test/registered/cpu/test_bmm.py b/test/registered/cpu/test_bmm.py
index f1399e9a5..42eedda23 100644
--- a/test/registered/cpu/test_bmm.py
+++ b/test/registered/cpu/test_bmm.py
@@ -11,7 +11,7 @@ from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.cpu_test_utils import precision
 from sglang.test.test_utils import CustomTestCase
 
-register_cpu_ci(est_time=13, suite="base-b-test-cpu")
+register_cpu_ci(est_time=13, suite="stage-a-test-cpu-intel")
 
 torch.manual_seed(1234)
 
diff --git a/test/registered/cpu/test_causal_conv1d.py b/test/registered/cpu/test_causal_conv1d.py
index 58e8454d5..70e89ac59 100644
--- a/test/registered/cpu/test_causal_conv1d.py
+++ b/test/registered/cpu/test_causal_conv1d.py
@@ -9,7 +9,7 @@ from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.cpu_test_utils import parametrize, precision
 from sglang.test.test_utils import CustomTestCase
 
-register_cpu_ci(est_time=7, suite="base-b-test-cpu")
+register_cpu_ci(est_time=7, suite="stage-a-test-cpu-intel")
 
 causal_conv1d_weight_pack = torch.ops.sgl_kernel.causal_conv1d_weight_pack
 causal_conv1d_fwd = torch.ops.sgl_kernel.causal_conv1d_fwd_cpu
diff --git a/test/registered/cpu/test_comm.py b/test/registered/cpu/test_comm.py
index 4cc175bf9..dc08bb92d 100644
--- a/test/registered/cpu/test_comm.py
+++ b/test/registered/cpu/test_comm.py
@@ -13,7 +13,7 @@ import torch.multiprocessing as mp
 from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.test_utils import CustomTestCase, find_available_port
 
-register_cpu_ci(est_time=43, suite="base-b-test-cpu")
+register_cpu_ci(est_time=43, suite="stage-a-test-cpu-intel")
 
 
 def run_distributed_test(rank, world_size, master_port, output_writer, fn):
diff --git a/test/registered/cpu/test_cpu_graph.py b/test/registered/cpu/test_cpu_graph.py
index 9d624c597..55be85eb7 100644
--- a/test/registered/cpu/test_cpu_graph.py
+++ b/test/registered/cpu/test_cpu_graph.py
@@ -21,7 +21,7 @@ from sglang.test.test_utils import (
     popen_launch_server,
 )
 
-register_cpu_ci(est_time=315, suite="base-b-tp-test-cpu")
+register_cpu_ci(est_time=315, suite="stage-a-tp-test-cpu-intel")
 
 
 class TestCPUGraph(CustomTestCase):
diff --git a/test/registered/cpu/test_decode.py b/test/registered/cpu/test_decode.py
index 15db49373..5f4b68c7d 100644
--- a/test/registered/cpu/test_decode.py
+++ b/test/registered/cpu/test_decode.py
@@ -7,7 +7,7 @@ from torch.nn.functional import scaled_dot_product_attention
 from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.test_utils import CustomTestCase
 
-register_cpu_ci(est_time=10, suite="base-b-test-cpu")
+register_cpu_ci(est_time=10, suite="stage-a-test-cpu-intel")
 register_cpu_ci(est_time=10, suite="base-b-test-cpu-arm64")
 
 torch.manual_seed(1234)
diff --git a/test/registered/cpu/test_extend.py b/test/registered/cpu/test_extend.py
index 40807a920..255a3f0e8 100644
--- a/test/registered/cpu/test_extend.py
+++ b/test/registered/cpu/test_extend.py
@@ -7,7 +7,7 @@ from torch.nn.functional import scaled_dot_product_attention
 from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.test_utils import CustomTestCase
 
-register_cpu_ci(est_time=7, suite="base-b-test-cpu")
+register_cpu_ci(est_time=7, suite="stage-a-test-cpu-intel")
 
 torch.manual_seed(1234)
 
diff --git a/test/registered/cpu/test_flash_attn.py b/test/registered/cpu/test_flash_attn.py
index 1ac9f7a5a..536b77402 100644
--- a/test/registered/cpu/test_flash_attn.py
+++ b/test/registered/cpu/test_flash_attn.py
@@ -8,7 +8,7 @@ from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.cpu_test_utils import parametrize, precision
 from sglang.test.test_utils import CustomTestCase
 
-register_cpu_ci(est_time=54, suite="base-b-test-cpu")
+register_cpu_ci(est_time=54, suite="stage-a-test-cpu-intel")
 
 flash_attn_varlen_func = torch.ops.sgl_kernel.flash_attn_varlen_func
 
diff --git a/test/registered/cpu/test_gemm.py b/test/registered/cpu/test_gemm.py
index 3011f6374..d3057c0a2 100644
--- a/test/registered/cpu/test_gemm.py
+++ b/test/registered/cpu/test_gemm.py
@@ -18,7 +18,7 @@ from sglang.test.cpu_test_utils import (
 )
 from sglang.test.test_utils import CustomTestCase
 
-register_cpu_ci(est_time=9, suite="base-b-test-cpu")
+register_cpu_ci(est_time=9, suite="stage-a-test-cpu-intel")
 
 torch.manual_seed(1234)
 
diff --git a/test/registered/cpu/test_intel_amx_attention_backend_a.py b/test/registered/cpu/test_intel_amx_attention_backend_a.py
index 7fe178812..3854dbdf5 100644
--- a/test/registered/cpu/test_intel_amx_attention_backend_a.py
+++ b/test/registered/cpu/test_intel_amx_attention_backend_a.py
@@ -21,7 +21,7 @@ from sglang.test.test_utils import (
     popen_launch_server,
 )
 
-register_cpu_ci(est_time=685, suite="base-b-tp-test-cpu")
+register_cpu_ci(est_time=685, suite="stage-a-tp-test-cpu-intel")
 
 
 class TestIntelAMXAttnBackend(CustomTestCase):
diff --git a/test/registered/cpu/test_intel_amx_attention_backend_b.py b/test/registered/cpu/test_intel_amx_attention_backend_b.py
index 9c088237c..de59dd9a4 100644
--- a/test/registered/cpu/test_intel_amx_attention_backend_b.py
+++ b/test/registered/cpu/test_intel_amx_attention_backend_b.py
@@ -14,7 +14,7 @@ from sglang.test.test_utils import (
     intel_amx_benchmark,
 )
 
-register_cpu_ci(est_time=47, suite="base-b-test-cpu")
+register_cpu_ci(est_time=47, suite="stage-a-test-cpu-intel")
 
 
 class TestIntelAMXAttnBackendQuant(CustomTestCase):
diff --git a/test/registered/cpu/test_intel_amx_attention_backend_c.py b/test/registered/cpu/test_intel_amx_attention_backend_c.py
index 3445b2d09..5fde261bf 100644
--- a/test/registered/cpu/test_intel_amx_attention_backend_c.py
+++ b/test/registered/cpu/test_intel_amx_attention_backend_c.py
@@ -14,7 +14,7 @@ from sglang.test.test_utils import (
     intel_amx_benchmark,
 )
 
-register_cpu_ci(est_time=477, suite="base-b-tp-test-cpu")
+register_cpu_ci(est_time=477, suite="stage-a-tp-test-cpu-intel")
 
 
 class TestIntelAMXAttnBackendQuant(CustomTestCase):
diff --git a/test/registered/cpu/test_mamba.py b/test/registered/cpu/test_mamba.py
index baaee7ec5..0f2148a51 100644
--- a/test/registered/cpu/test_mamba.py
+++ b/test/registered/cpu/test_mamba.py
@@ -9,7 +9,7 @@ from torch.nn.functional import softplus
 from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.cpu_test_utils import precision
 
-register_cpu_ci(est_time=7, suite="base-b-test-cpu")
+register_cpu_ci(est_time=7, suite="stage-a-test-cpu-intel")
 
 torch.manual_seed(1234)
 
diff --git a/test/registered/cpu/test_mla.py b/test/registered/cpu/test_mla.py
index fb17933aa..280a10b09 100644
--- a/test/registered/cpu/test_mla.py
+++ b/test/registered/cpu/test_mla.py
@@ -8,7 +8,7 @@ from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.cpu_test_utils import precision
 from sglang.test.test_utils import CustomTestCase
 
-register_cpu_ci(est_time=8, suite="base-b-test-cpu")
+register_cpu_ci(est_time=8, suite="stage-a-test-cpu-intel")
 
 torch.manual_seed(1234)
 
diff --git a/test/registered/cpu/test_moe.py b/test/registered/cpu/test_moe.py
index 4811f51f3..46765e243 100644
--- a/test/registered/cpu/test_moe.py
+++ b/test/registered/cpu/test_moe.py
@@ -34,7 +34,7 @@ from sglang.test.cpu_test_utils import (
     unpack_and_dequant_awq,
 )
 
-register_cpu_ci(est_time=7, suite="base-b-test-cpu")
+register_cpu_ci(est_time=7, suite="stage-a-test-cpu-intel")
 
 
 def run_fused_experts(
diff --git a/test/registered/cpu/test_norm.py b/test/registered/cpu/test_norm.py
index f03825d09..cd8c340ed 100644
--- a/test/registered/cpu/test_norm.py
+++ b/test/registered/cpu/test_norm.py
@@ -8,7 +8,7 @@ import torch
 from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.cpu_test_utils import make_non_contiguous, precision
 
-register_cpu_ci(est_time=5, suite="base-b-test-cpu")
+register_cpu_ci(est_time=5, suite="stage-a-test-cpu-intel")
 register_cpu_ci(est_time=10, suite="base-b-test-cpu-arm64")
 
 torch.manual_seed(1234)
diff --git a/test/registered/cpu/test_qkv_proj_with_rope.py b/test/registered/cpu/test_qkv_proj_with_rope.py
index 2c0cda2e9..29fcdf6e1 100644
--- a/test/registered/cpu/test_qkv_proj_with_rope.py
+++ b/test/registered/cpu/test_qkv_proj_with_rope.py
@@ -14,7 +14,7 @@ from sglang.test.cpu_test_utils import (
 )
 from sglang.test.test_utils import CustomTestCase
 
-register_cpu_ci(est_time=7, suite="base-b-test-cpu")
+register_cpu_ci(est_time=7, suite="stage-a-test-cpu-intel")
 
 convert_weight_packed = torch.ops.sgl_kernel.convert_weight_packed
 qkv_proj_with_rope = torch.ops.sgl_kernel.qkv_proj_with_rope
diff --git a/test/registered/cpu/test_qwen3.py b/test/registered/cpu/test_qwen3.py
index 202bdf298..f0cb043e7 100644
--- a/test/registered/cpu/test_qwen3.py
+++ b/test/registered/cpu/test_qwen3.py
@@ -8,7 +8,7 @@ from sglang.srt.utils import is_host_cpu_arm64
 from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.cpu_test_utils import precision
 
-register_cpu_ci(est_time=5, suite="base-b-test-cpu")
+register_cpu_ci(est_time=5, suite="stage-a-test-cpu-intel")
 register_cpu_ci(est_time=10, suite="base-b-test-cpu-arm64")
 
 torch.manual_seed(1234)
diff --git a/test/registered/cpu/test_rank_consensus_checker.py b/test/registered/cpu/test_rank_consensus_checker.py
index 0bc7b421e..773c4b3ff 100644
--- a/test/registered/cpu/test_rank_consensus_checker.py
+++ b/test/registered/cpu/test_rank_consensus_checker.py
@@ -25,7 +25,7 @@ from sglang.srt.utils.rank_consensus_checker import (
 from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.test_utils import CustomTestCase, find_available_port
 
-register_cpu_ci(est_time=193, suite="base-b-test-cpu")
+register_cpu_ci(est_time=193, suite="stage-a-test-cpu-intel")
 
 
 def run_distributed_test(
diff --git a/test/registered/cpu/test_request_decompression.py b/test/registered/cpu/test_request_decompression.py
index 003a57d6d..06a7ce556 100644
--- a/test/registered/cpu/test_request_decompression.py
+++ b/test/registered/cpu/test_request_decompression.py
@@ -8,7 +8,7 @@ from sglang.srt.entrypoints.http_request_decompression import (
 )
 from sglang.test.ci.ci_register import register_cpu_ci
 
-register_cpu_ci(est_time=5, suite="base-b-test-cpu")
+register_cpu_ci(est_time=5, suite="stage-a-test-cpu-intel")
 
 PAYLOAD = b'{"text":"hello world","n":7}'
 COMPRESSED = zstandard.ZstdCompressor().compress(PAYLOAD)
diff --git a/test/registered/cpu/test_request_headers.py b/test/registered/cpu/test_request_headers.py
index 5e0504b08..392a4f57d 100644
--- a/test/registered/cpu/test_request_headers.py
+++ b/test/registered/cpu/test_request_headers.py
@@ -7,7 +7,7 @@ from starlette.datastructures import Headers
 from sglang.srt.entrypoints.request_headers import apply_header_overrides
 from sglang.test.ci.ci_register import register_cpu_ci
 
-register_cpu_ci(est_time=4, suite="base-b-test-cpu")
+register_cpu_ci(est_time=4, suite="stage-a-test-cpu-intel")
 
 
 def _obj():
diff --git a/test/registered/cpu/test_rope.py b/test/registered/cpu/test_rope.py
index d5b24e40a..0425e09d3 100644
--- a/test/registered/cpu/test_rope.py
+++ b/test/registered/cpu/test_rope.py
@@ -17,7 +17,7 @@ from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.cpu_test_utils import precision
 from sglang.test.test_utils import CustomTestCase
 
-register_cpu_ci(est_time=33, suite="base-b-test-cpu")
+register_cpu_ci(est_time=33, suite="stage-a-test-cpu-intel")
 register_cpu_ci(est_time=10, suite="base-b-test-cpu-arm64")
 
 torch.manual_seed(1234)
diff --git a/test/registered/cpu/test_server_args_backend.py b/test/registered/cpu/test_server_args_backend.py
index 5c710fa84..52d8aadaa 100644
--- a/test/registered/cpu/test_server_args_backend.py
+++ b/test/registered/cpu/test_server_args_backend.py
@@ -10,7 +10,7 @@ from sglang.srt.arg_groups.validation_hook import validate_ib_devices
 from sglang.srt.server_args import ServerArgs
 from sglang.test.ci.ci_register import register_cpu_ci
 
-register_cpu_ci(est_time=6, suite="base-b-test-cpu")
+register_cpu_ci(est_time=6, suite="stage-a-test-cpu-intel")
 register_cpu_ci(est_time=10, suite="base-b-test-cpu-arm64")
 
 
diff --git a/test/registered/cpu/test_shared_expert.py b/test/registered/cpu/test_shared_expert.py
index a2a8bbe41..3aecf410a 100644
--- a/test/registered/cpu/test_shared_expert.py
+++ b/test/registered/cpu/test_shared_expert.py
@@ -20,7 +20,7 @@ from sglang.test.cpu_test_utils import (
 )
 from sglang.test.test_utils import CustomTestCase
 
-register_cpu_ci(est_time=8, suite="base-b-test-cpu")
+register_cpu_ci(est_time=8, suite="stage-a-test-cpu-intel")
 
 torch.manual_seed(1234)
 
diff --git a/test/registered/cpu/test_spec_eagle_cpu.py b/test/registered/cpu/test_spec_eagle_cpu.py
index 06817361d..b0e70b3e9 100644
--- a/test/registered/cpu/test_spec_eagle_cpu.py
+++ b/test/registered/cpu/test_spec_eagle_cpu.py
@@ -20,7 +20,7 @@ from sglang.test.server_fixtures.spec_eagle_fixture import EagleLlama2Base
 # Measured 780s all-green on a 40-core GNR socket (1 launch + 18 methods).
 register_cpu_ci(
     est_time=800,
-    suite="base-b-test-cpu",
+    suite="stage-a-test-cpu-intel",
     disabled="EagleLlama2Base needs gated meta-llama/Llama-2-7b-chat-hf",
 )
 
diff --git a/test/registered/cpu/test_spec_eagle_parity_cpu.py b/test/registered/cpu/test_spec_eagle_parity_cpu.py
index 724e702d9..1f446f2fa 100644
--- a/test/registered/cpu/test_spec_eagle_parity_cpu.py
+++ b/test/registered/cpu/test_spec_eagle_parity_cpu.py
@@ -9,7 +9,7 @@ from sglang.test.server_fixtures.spec_eagle_fixture import Eagle3Base
 # (CUDA sibling: 360); tune from CI TIMINGS once it has run there.
 register_cpu_ci(
     est_time=480,
-    suite="base-b-test-cpu",
+    suite="stage-a-test-cpu-intel",
     disabled="EAGLE3 numerical parity mismatches on CPU intel_amx",
 )
 
diff --git a/test/registered/cpu/test_spec_eagle_topk_cpu.py b/test/registered/cpu/test_spec_eagle_topk_cpu.py
index bc1949c5b..ecd02af4b 100644
--- a/test/registered/cpu/test_spec_eagle_topk_cpu.py
+++ b/test/registered/cpu/test_spec_eagle_topk_cpu.py
@@ -21,7 +21,7 @@ from sglang.test.server_fixtures.spec_eagle_fixture import EagleLlama2Base
 # Measured 830s all-green on a 40-core GNR socket (1 launch + 14 methods).
 register_cpu_ci(
     est_time=850,
-    suite="base-b-test-cpu",
+    suite="stage-a-test-cpu-intel",
     disabled="EagleLlama2Base needs gated meta-llama/Llama-2-7b-chat-hf",
 )
 
diff --git a/test/registered/cpu/test_spec_kernels.py b/test/registered/cpu/test_spec_kernels.py
index bbb9131b2..8808a4b91 100644
--- a/test/registered/cpu/test_spec_kernels.py
+++ b/test/registered/cpu/test_spec_kernels.py
@@ -9,7 +9,7 @@ from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.cpu_test_utils import precision
 from sglang.test.test_utils import CustomTestCase
 
-register_cpu_ci(est_time=6, suite="base-b-test-cpu")
+register_cpu_ci(est_time=6, suite="stage-a-test-cpu-intel")
 
 
 def _topk1_chain_inputs(bs, num_steps):
diff --git a/test/registered/cpu/test_store_cache.py b/test/registered/cpu/test_store_cache.py
index 4d09da3be..3a010bee4 100644
--- a/test/registered/cpu/test_store_cache.py
+++ b/test/registered/cpu/test_store_cache.py
@@ -11,7 +11,7 @@ from sglang.srt.layers.quantization.fp4_kv_cache_quant_method import (
 from sglang.srt.mem_cache.memory_pool import MHATokenToKVPool
 from sglang.test.ci.ci_register import register_cpu_ci
 
-register_cpu_ci(est_time=14, suite="base-b-test-cpu")
+register_cpu_ci(est_time=14, suite="stage-a-test-cpu-intel")
 
 torch.manual_seed(42)
 
diff --git a/test/registered/cpu/test_subblock_sparse_attention.py b/test/registered/cpu/test_subblock_sparse_attention.py
index dace72d5b..ecfcf926f 100644
--- a/test/registered/cpu/test_subblock_sparse_attention.py
+++ b/test/registered/cpu/test_subblock_sparse_attention.py
@@ -43,7 +43,7 @@ from sglang.multimodal_gen.runtime.platforms.cuda import (
 from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.test_utils import CustomTestCase
 
-register_cpu_ci(est_time=9, suite="base-b-test-cpu")
+register_cpu_ci(est_time=9, suite="stage-a-test-cpu-intel")
 
 
 class TestSubBlockSparseAttentionDispatch(CustomTestCase):
diff --git a/test/registered/cpu/test_topk.py b/test/registered/cpu/test_topk.py
index e3f469b40..d4da6ee47 100644
--- a/test/registered/cpu/test_topk.py
+++ b/test/registered/cpu/test_topk.py
@@ -15,7 +15,7 @@ from sglang.srt.models.llama4 import Llama4MoE
 from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.test_utils import CustomTestCase
 
-register_cpu_ci(est_time=30, suite="base-b-test-cpu")
+register_cpu_ci(est_time=30, suite="stage-a-test-cpu-intel")
 register_cpu_ci(est_time=10, suite="base-b-test-cpu-arm64")
 
 
diff --git a/test/registered/debug_utils/test_crash_dump.py b/test/registered/debug_utils/test_crash_dump.py
index 7a825d4c5..d83244150 100644
--- a/test/registered/debug_utils/test_crash_dump.py
+++ b/test/registered/debug_utils/test_crash_dump.py
@@ -23,7 +23,7 @@ from sglang.test.test_utils import (
 
 register_cuda_ci(est_time=70, stage="nightly", runner_config="1-gpu-large")
 register_amd_ci(est_time=40, suite="nightly-amd-1-gpu", nightly=True)
-register_cpu_ci(est_time=48, suite="base-c-test-cpu")
+register_cpu_ci(est_time=48, suite="stage-b-test-cpu-intel")
 
 
 class TestCrashDump(CustomTestCase):
diff --git a/test/registered/debug_utils/test_soft_watchdog.py b/test/registered/debug_utils/test_soft_watchdog.py
index fa2f0c6b2..4c828bd09 100644
--- a/test/registered/debug_utils/test_soft_watchdog.py
+++ b/test/registered/debug_utils/test_soft_watchdog.py
@@ -19,7 +19,7 @@ from sglang.test.test_utils import (
 
 register_cuda_ci(est_time=240, stage="nightly", runner_config="1-gpu-large")
 register_amd_ci(est_time=120, suite="nightly-amd-1-gpu", nightly=True)
-register_cpu_ci(est_time=237, suite="base-c-test-cpu")
+register_cpu_ci(est_time=237, suite="stage-b-test-cpu-intel")
 
 
 class BaseTestSoftWatchdog:
diff --git a/test/registered/function_call/test_kimik2_detector.py b/test/registered/function_call/test_kimik2_detector.py
index 3b182c8f3..4a4dc90ef 100644
--- a/test/registered/function_call/test_kimik2_detector.py
+++ b/test/registered/function_call/test_kimik2_detector.py
@@ -12,7 +12,7 @@ from sglang.srt.parser.reasoning_parser import KimiK2Detector as KimiK2Reasoning
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(5, "base-a-test-cpu")
-register_cpu_ci(est_time=4, suite="base-c-test-cpu")
+register_cpu_ci(est_time=4, suite="stage-b-test-cpu-intel")
 
 
 def _make_tool(name, parameters=None):
diff --git a/test/registered/input_embedding/test_input_embeddings.py b/test/registered/input_embedding/test_input_embeddings.py
index 115d548aa..616d5ea68 100644
--- a/test/registered/input_embedding/test_input_embeddings.py
+++ b/test/registered/input_embedding/test_input_embeddings.py
@@ -22,7 +22,7 @@ from sglang.test.test_utils import (
 
 register_cuda_ci(est_time=42, stage="base-b", runner_config="1-gpu-small")
 register_amd_ci(est_time=38, suite="stage-b-test-1-gpu-small-amd")
-register_cpu_ci(est_time=58, suite="base-c-test-cpu")
+register_cpu_ci(est_time=58, suite="stage-b-test-cpu-intel")
 
 
 class TestInputEmbeds(CustomTestCase):
diff --git a/test/registered/input_embedding/test_input_embeds_chunked.py b/test/registered/input_embedding/test_input_embeds_chunked.py
index 5e1ce4a5d..bdab8c8fa 100644
--- a/test/registered/input_embedding/test_input_embeds_chunked.py
+++ b/test/registered/input_embedding/test_input_embeds_chunked.py
@@ -36,7 +36,7 @@ from sglang.test.test_utils import (
 
 register_cuda_ci(est_time=43, stage="base-b", runner_config="1-gpu-small")
 register_amd_ci(est_time=43, suite="stage-b-test-1-gpu-small-amd")
-register_cpu_ci(est_time=54, suite="base-c-test-cpu")
+register_cpu_ci(est_time=54, suite="stage-b-test-cpu-intel")
 
 CHUNKED_PREFILL_SIZE = 256
 
diff --git a/test/registered/kv_canary/test_e2e_base.py b/test/registered/kv_canary/test_e2e_base.py
index 838ba7ad2..459053307 100644
--- a/test/registered/kv_canary/test_e2e_base.py
+++ b/test/registered/kv_canary/test_e2e_base.py
@@ -8,7 +8,7 @@ from sglang.test.kv_canary.e2e_base import (
 )
 from sglang.test.test_utils import CustomTestCase
 
-register_cpu_ci(est_time=8, suite="base-c-test-cpu")
+register_cpu_ci(est_time=8, suite="stage-b-test-cpu-intel")
 
 
 class TestCanaryE2EBase(CustomTestCase):
diff --git a/test/registered/kv_canary/test_self_unit_e2e_base.py b/test/registered/kv_canary/test_self_unit_e2e_base.py
index b052f22c0..a8e8f5d7d 100644
--- a/test/registered/kv_canary/test_self_unit_e2e_base.py
+++ b/test/registered/kv_canary/test_self_unit_e2e_base.py
@@ -8,7 +8,7 @@ from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.kv_canary.e2e_base import CanaryE2EBase
 from sglang.test.test_utils import CustomTestCase
 
-register_cpu_ci(est_time=6, suite="base-c-test-cpu")
+register_cpu_ci(est_time=6, suite="stage-b-test-cpu-intel")
 
 
 _GOOD_LINE: str = SwaDivergenceLog(
diff --git a/test/registered/lora/test_lora_openai_api.py b/test/registered/lora/test_lora_openai_api.py
index 1897c0113..7df80bd74 100644
--- a/test/registered/lora/test_lora_openai_api.py
+++ b/test/registered/lora/test_lora_openai_api.py
@@ -13,7 +13,7 @@ from sglang.srt.server_args import ServerArgs
 from sglang.test.ci.ci_register import register_amd_ci, register_cpu_ci
 
 register_amd_ci(est_time=30, suite="nightly-amd-1-gpu", nightly=True)
-register_cpu_ci(est_time=7, suite="base-c-test-cpu")
+register_cpu_ci(est_time=7, suite="stage-b-test-cpu-intel")
 
 
 class MockTokenizerManager:
diff --git a/test/registered/model_loading/test_external_models.py b/test/registered/model_loading/test_external_models.py
index f72d70e3b..9f54fd843 100644
--- a/test/registered/model_loading/test_external_models.py
+++ b/test/registered/model_loading/test_external_models.py
@@ -11,7 +11,7 @@ from sglang.test.test_utils import CustomTestCase
 
 register_cuda_ci(est_time=29, stage="base-b", runner_config="1-gpu-small")
 register_amd_ci(est_time=45, suite="stage-b-test-1-gpu-small-amd")
-register_cpu_ci(est_time=32, suite="base-c-test-cpu")
+register_cpu_ci(est_time=32, suite="stage-b-test-cpu-intel")
 
 
 class TestExternalModels(CustomTestCase):
diff --git a/test/registered/models_e2e/test_transformers_backend_eval.py b/test/registered/models_e2e/test_transformers_backend_eval.py
index 5da8d2551..ab1cd08dd 100644
--- a/test/registered/models_e2e/test_transformers_backend_eval.py
+++ b/test/registered/models_e2e/test_transformers_backend_eval.py
@@ -13,7 +13,7 @@ from sglang.test.server_fixtures.default_fixture import DefaultServerBase
 
 register_cuda_ci(est_time=48, stage="base-b", runner_config="1-gpu-small")
 register_amd_ci(est_time=48, suite="stage-b-test-1-gpu-small-amd")
-register_cpu_ci(est_time=206, suite="base-c-test-cpu")
+register_cpu_ci(est_time=206, suite="stage-b-test-cpu-intel")
 
 
 class TestTransformersBackendEval(DefaultServerBase):
diff --git a/test/registered/moe/test_hash_topk.py b/test/registered/moe/test_hash_topk.py
index c8e14c467..1c5658313 100644
--- a/test/registered/moe/test_hash_topk.py
+++ b/test/registered/moe/test_hash_topk.py
@@ -15,7 +15,7 @@ from sglang.srt.runtime_context import get_parallel
 from sglang.srt.server_args import ServerArgs, set_global_server_args_for_scheduler
 from sglang.test.ci.ci_register import register_cpu_ci
 
-register_cpu_ci(est_time=6, suite="base-b-test-cpu")
+register_cpu_ci(est_time=6, suite="stage-a-test-cpu-intel")
 
 
 @pytest.fixture(autouse=True)
diff --git a/test/registered/observability/test_priority_metrics.py b/test/registered/observability/test_priority_metrics.py
index 5a7104813..593d8b4eb 100644
--- a/test/registered/observability/test_priority_metrics.py
+++ b/test/registered/observability/test_priority_metrics.py
@@ -26,7 +26,7 @@ register_cuda_ci(
     runner_config="1-gpu-small",
 )
 register_amd_ci(est_time=60, suite="stage-b-test-1-gpu-small-amd")
-register_cpu_ci(est_time=49, suite="base-c-test-cpu")
+register_cpu_ci(est_time=49, suite="stage-b-test-cpu-intel")
 
 _MODEL_NAME = "Qwen/Qwen3-0.6B"
 
diff --git a/test/registered/openai_server/function_call/test_anthropic_tool_use.py b/test/registered/openai_server/function_call/test_anthropic_tool_use.py
index 560b4ef9c..2a52e4378 100644
--- a/test/registered/openai_server/function_call/test_anthropic_tool_use.py
+++ b/test/registered/openai_server/function_call/test_anthropic_tool_use.py
@@ -31,7 +31,7 @@ from sglang.test.test_utils import (
 
 register_cuda_ci(est_time=50, stage="base-b", runner_config="1-gpu-large")
 register_amd_ci(est_time=140, suite="stage-b-test-1-gpu-small-amd")
-register_cpu_ci(est_time=54, suite="base-c-test-cpu")
+register_cpu_ci(est_time=54, suite="stage-b-test-cpu-intel")
 
 # System message to guide Llama3.2 to produce proper tool call format
 SYSTEM_MESSAGE = (
diff --git a/test/registered/openai_server/validation/test_large_max_new_tokens.py b/test/registered/openai_server/validation/test_large_max_new_tokens.py
index 5f674066e..4138d336f 100644
--- a/test/registered/openai_server/validation/test_large_max_new_tokens.py
+++ b/test/registered/openai_server/validation/test_large_max_new_tokens.py
@@ -28,7 +28,7 @@ from sglang.test.test_utils import (
 
 register_cuda_ci(est_time=58, stage="base-b", runner_config="1-gpu-large")
 register_amd_ci(est_time=41, suite="stage-b-test-1-gpu-small-amd")
-register_cpu_ci(est_time=101, suite="base-c-test-cpu")
+register_cpu_ci(est_time=101, suite="stage-b-test-cpu-intel")
 
 
 class TestLargeMaxNewTokens(CustomTestCase):
diff --git a/test/registered/openai_server/validation/test_matched_stop.py b/test/registered/openai_server/validation/test_matched_stop.py
index be2853138..e8345c504 100644
--- a/test/registered/openai_server/validation/test_matched_stop.py
+++ b/test/registered/openai_server/validation/test_matched_stop.py
@@ -16,7 +16,7 @@ from sglang.test.test_utils import (
 
 register_cuda_ci(est_time=52, stage="base-b", runner_config="1-gpu-small")
 register_amd_ci(est_time=60, suite="stage-b-test-1-gpu-small-amd")
-register_cpu_ci(est_time=83, suite="base-c-test-cpu")
+register_cpu_ci(est_time=83, suite="stage-b-test-cpu-intel")
 
 
 class TestMatchedStop(CustomTestCase, MatchedStopMixin):
diff --git a/test/registered/prefill_only/test_openai_embedding.py b/test/registered/prefill_only/test_openai_embedding.py
index a5ff4ef23..20f01951f 100644
--- a/test/registered/prefill_only/test_openai_embedding.py
+++ b/test/registered/prefill_only/test_openai_embedding.py
@@ -19,7 +19,7 @@ from sglang.test.test_utils import (
 
 register_cuda_ci(est_time=91, stage="base-b", runner_config="1-gpu-small")
 register_amd_ci(est_time=141, suite="stage-b-test-1-gpu-small-amd")
-register_cpu_ci(est_time=91, suite="base-c-test-cpu")
+register_cpu_ci(est_time=91, suite="stage-b-test-cpu-intel")
 
 
 class TestOpenAIEmbedding(CustomTestCase):
diff --git a/test/registered/quant/test_quant_config_parsing.py b/test/registered/quant/test_quant_config_parsing.py
index a27e56619..b16ae0b33 100644
--- a/test/registered/quant/test_quant_config_parsing.py
+++ b/test/registered/quant/test_quant_config_parsing.py
@@ -6,7 +6,7 @@ from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.test_utils import CustomTestCase
 
 register_cpu_ci(est_time=15, suite="base-a-test-cpu")
-register_cpu_ci(est_time=8, suite="base-c-test-cpu")
+register_cpu_ci(est_time=8, suite="stage-b-test-cpu-intel")
 
 
 class TestQuantLogString(CustomTestCase):
diff --git a/test/registered/radix_cache/test_radix_attention.py b/test/registered/radix_cache/test_radix_attention.py
index 0f7172f4b..bdce80433 100644
--- a/test/registered/radix_cache/test_radix_attention.py
+++ b/test/registered/radix_cache/test_radix_attention.py
@@ -20,7 +20,7 @@ from sglang.test.test_utils import (
 # RadixAttention server integration tests
 register_cuda_ci(est_time=100, stage="base-b", runner_config="1-gpu-small")
 register_amd_ci(est_time=100, suite="stage-b-test-1-gpu-small-amd")
-register_cpu_ci(est_time=405, suite="base-c-test-cpu")
+register_cpu_ci(est_time=405, suite="stage-b-test-cpu-intel")
 
 
 class TestRadixCacheFCFS(CustomTestCase):
diff --git a/test/registered/radix_cache/test_radix_cache_hit.py b/test/registered/radix_cache/test_radix_cache_hit.py
index 7ad093064..a62eba9ef 100644
--- a/test/registered/radix_cache/test_radix_cache_hit.py
+++ b/test/registered/radix_cache/test_radix_cache_hit.py
@@ -17,7 +17,7 @@ from sglang.test.test_utils import (
 
 register_cuda_ci(est_time=55, stage="base-b", runner_config="1-gpu-small")
 register_amd_ci(est_time=55, suite="stage-b-test-1-gpu-small-amd")
-register_cpu_ci(est_time=113, suite="base-c-test-cpu")
+register_cpu_ci(est_time=113, suite="stage-b-test-cpu-intel")
 
 MODEL = DEFAULT_SMALL_MODEL_NAME_FOR_TEST
 
diff --git a/test/registered/radix_cache/test_swa_radix_cache_kl.py b/test/registered/radix_cache/test_swa_radix_cache_kl.py
index e34372221..918138fe5 100644
--- a/test/registered/radix_cache/test_swa_radix_cache_kl.py
+++ b/test/registered/radix_cache/test_swa_radix_cache_kl.py
@@ -7,7 +7,7 @@ from sglang.test.server_fixtures.default_fixture import DefaultServerBase
 MODEL = "openai/gpt-oss-20b"
 
 register_cuda_ci(est_time=151, stage="base-b", runner_config="1-gpu-large")
-register_cpu_ci(est_time=1602, suite="base-c-test-cpu")
+register_cpu_ci(est_time=1602, suite="stage-b-test-cpu-intel")
 
 
 class TestSWARadixCacheKL(KLDivergenceMixin, DefaultServerBase):
diff --git a/test/registered/rl/test_pause_generation_tensor_consistency.py b/test/registered/rl/test_pause_generation_tensor_consistency.py
index c82c1976c..7871e812e 100644
--- a/test/registered/rl/test_pause_generation_tensor_consistency.py
+++ b/test/registered/rl/test_pause_generation_tensor_consistency.py
@@ -10,7 +10,7 @@ from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.test_utils import CustomTestCase
 
 register_cpu_ci(est_time=7, suite="base-a-test-cpu")
-register_cpu_ci(est_time=8, suite="base-c-test-cpu")
+register_cpu_ci(est_time=8, suite="stage-b-test-cpu-intel")
 
 
 # ---------------------------------------------------------------------------
diff --git a/test/registered/scheduler/test_abort_with_metrics.py b/test/registered/scheduler/test_abort_with_metrics.py
index b35fb210e..b78d2a2b6 100644
--- a/test/registered/scheduler/test_abort_with_metrics.py
+++ b/test/registered/scheduler/test_abort_with_metrics.py
@@ -19,7 +19,7 @@ from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.test_utils import CustomTestCase
 
 register_cpu_ci(est_time=7, suite="base-a-test-cpu")
-register_cpu_ci(est_time=8, suite="base-c-test-cpu")
+register_cpu_ci(est_time=8, suite="stage-b-test-cpu-intel")
 
 _HTTP_SCOPE = {
     "type": "http",
diff --git a/test/registered/scheduler/test_routing_key_scheduling.py b/test/registered/scheduler/test_routing_key_scheduling.py
index 7c5e4256d..7b4d7bf37 100644
--- a/test/registered/scheduler/test_routing_key_scheduling.py
+++ b/test/registered/scheduler/test_routing_key_scheduling.py
@@ -22,7 +22,7 @@ from sglang.test.test_utils import (
 
 register_cuda_ci(est_time=50, stage="weekly", runner_config="1-gpu-large")
 register_amd_ci(est_time=120, suite="nightly-amd-1-gpu", nightly=True)
-register_cpu_ci(est_time=55, suite="base-c-test-cpu")
+register_cpu_ci(est_time=55, suite="stage-b-test-cpu-intel")
 
 
 class TestRoutingKeyScheduling(CustomTestCase):
diff --git a/test/registered/sessions/test_streaming_session_swa_extra.py b/test/registered/sessions/test_streaming_session_swa_extra.py
index 52c8cc13f..ba2dbda46 100644
--- a/test/registered/sessions/test_streaming_session_swa_extra.py
+++ b/test/registered/sessions/test_streaming_session_swa_extra.py
@@ -14,7 +14,7 @@ from sglang.test.server_fixtures.streaming_session_fixture import (
 )
 
 register_cuda_ci(est_time=130, stage="extra-a", runner_config="1-gpu-large")
-register_cpu_ci(est_time=449, suite="base-c-test-cpu")
+register_cpu_ci(est_time=449, suite="stage-b-test-cpu-intel")
 
 
 class TestStreamingSessionSWARetractMixedChunk(
diff --git a/test/registered/unit/bench/test_mixed_prefix_gsm8k.py b/test/registered/unit/bench/test_mixed_prefix_gsm8k.py
index be32ae4be..aea618b40 100644
--- a/test/registered/unit/bench/test_mixed_prefix_gsm8k.py
+++ b/test/registered/unit/bench/test_mixed_prefix_gsm8k.py
@@ -11,7 +11,7 @@ from sglang.test.simple_eval_mixed_prefix_gsm8k import (
 )
 from sglang.test.test_utils import CustomTestCase
 
-register_cpu_ci(est_time=8, suite="base-c-test-cpu")
+register_cpu_ci(est_time=8, suite="stage-b-test-cpu-intel")
 
 
 def _write_synthetic_dataset(path: str, n: int) -> None:
diff --git a/test/registered/unit/bench/test_simple_eval_gsm8k.py b/test/registered/unit/bench/test_simple_eval_gsm8k.py
index 4c58c5333..8314691de 100644
--- a/test/registered/unit/bench/test_simple_eval_gsm8k.py
+++ b/test/registered/unit/bench/test_simple_eval_gsm8k.py
@@ -10,7 +10,7 @@ from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.run_eval import _run_sgl_eval, run_eval
 from sglang.test.test_utils import CustomTestCase
 
-register_cpu_ci(est_time=6, suite="base-b-test-cpu")
+register_cpu_ci(est_time=6, suite="stage-a-test-cpu-intel")
 
 
 def _write_fake_metrics(out_parent: Path, eval_name: str, payload: dict) -> None:
diff --git a/test/registered/unit/constrained/test_grammar_manager.py b/test/registered/unit/constrained/test_grammar_manager.py
index c1031e2b0..0d70b4008 100644
--- a/test/registered/unit/constrained/test_grammar_manager.py
+++ b/test/registered/unit/constrained/test_grammar_manager.py
@@ -34,7 +34,7 @@ from sglang.test.ci.ci_register import register_cpu_ci
 register_cpu_ci(2.0, "base-a-test-cpu")
 
 
-register_cpu_ci(est_time=5, suite="base-c-test-cpu")
+register_cpu_ci(est_time=5, suite="stage-b-test-cpu-intel")
 
 
 def _make_scheduler(grammar_backend_name="none", skip_tokenizer=False):
diff --git a/test/registered/unit/constrained/test_reasoner_grammar_backend.py b/test/registered/unit/constrained/test_reasoner_grammar_backend.py
index da1eaf219..ab32a2e66 100644
--- a/test/registered/unit/constrained/test_reasoner_grammar_backend.py
+++ b/test/registered/unit/constrained/test_reasoner_grammar_backend.py
@@ -18,7 +18,7 @@ from sglang.srt.parser.reasoning_parser import KimiK3Detector as KimiK3Reasoning
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(2.0, "base-a-test-cpu")
-register_cpu_ci(est_time=5, suite="base-c-test-cpu")
+register_cpu_ci(est_time=5, suite="stage-b-test-cpu-intel")
 
 
 class _DummyTokenizer:
diff --git a/test/registered/unit/constrained/test_utils.py b/test/registered/unit/constrained/test_utils.py
index 818a336f2..c1f589596 100644
--- a/test/registered/unit/constrained/test_utils.py
+++ b/test/registered/unit/constrained/test_utils.py
@@ -15,7 +15,7 @@ from sglang.srt.constrained.utils import is_legacy_structural_tag
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(1.0, "base-a-test-cpu")
-register_cpu_ci(est_time=5, suite="base-c-test-cpu")
+register_cpu_ci(est_time=5, suite="stage-b-test-cpu-intel")
 
 
 class TestIsLegacyStructuralTag(unittest.TestCase):
diff --git a/test/registered/unit/entrypoints/openai/test_serving_embedding.py b/test/registered/unit/entrypoints/openai/test_serving_embedding.py
index 40b7d292d..a9624fb81 100644
--- a/test/registered/unit/entrypoints/openai/test_serving_embedding.py
+++ b/test/registered/unit/entrypoints/openai/test_serving_embedding.py
@@ -59,7 +59,7 @@ from sglang.srt.managers.io_struct import EmbeddingReqInput
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(est_time=10, suite="base-a-test-cpu")
-register_cpu_ci(est_time=6, suite="base-c-test-cpu")
+register_cpu_ci(est_time=6, suite="stage-b-test-cpu-intel")
 
 
 # Mock TokenizerManager for embedding tests
diff --git a/test/registered/unit/function_call/test_function_call_parser.py b/test/registered/unit/function_call/test_function_call_parser.py
index ad7feaf24..5b14add1f 100644
--- a/test/registered/unit/function_call/test_function_call_parser.py
+++ b/test/registered/unit/function_call/test_function_call_parser.py
@@ -37,7 +37,7 @@ from sglang.srt.function_call.utils import get_schema_properties
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(est_time=20, suite="base-a-test-cpu")
-register_cpu_ci(est_time=70, suite="base-c-test-cpu")
+register_cpu_ci(est_time=70, suite="stage-b-test-cpu-intel")
 
 
 @functools.lru_cache(maxsize=None)
diff --git a/test/registered/unit/function_call/test_json_schema_constraint.py b/test/registered/unit/function_call/test_json_schema_constraint.py
index d53c607d3..d4e12b85f 100644
--- a/test/registered/unit/function_call/test_json_schema_constraint.py
+++ b/test/registered/unit/function_call/test_json_schema_constraint.py
@@ -19,7 +19,7 @@ from sglang.srt.function_call.utils import (
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(5, "base-a-test-cpu")
-register_cpu_ci(est_time=6, suite="base-c-test-cpu")
+register_cpu_ci(est_time=6, suite="stage-b-test-cpu-intel")
 
 
 class TestJsonSchemaConstraint(unittest.TestCase):
diff --git a/test/registered/unit/function_call/test_parallel_tool_calls.py b/test/registered/unit/function_call/test_parallel_tool_calls.py
index 6dbb6d2d6..fbaec23b1 100644
--- a/test/registered/unit/function_call/test_parallel_tool_calls.py
+++ b/test/registered/unit/function_call/test_parallel_tool_calls.py
@@ -24,7 +24,7 @@ from sglang.srt.function_call.json_array_parser import JsonArrayParser
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(5, "base-a-test-cpu")
-register_cpu_ci(est_time=4, suite="base-c-test-cpu")
+register_cpu_ci(est_time=4, suite="stage-b-test-cpu-intel")
 
 
 class TestParallelToolCalls(unittest.TestCase):
diff --git a/test/registered/unit/function_call/test_unknown_tool_name.py b/test/registered/unit/function_call/test_unknown_tool_name.py
index d07d2ed45..645ee0f83 100644
--- a/test/registered/unit/function_call/test_unknown_tool_name.py
+++ b/test/registered/unit/function_call/test_unknown_tool_name.py
@@ -10,7 +10,7 @@ from sglang.srt.function_call.core_types import StreamingParseResult
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(5, "base-a-test-cpu")
-register_cpu_ci(est_time=5, suite="base-c-test-cpu")
+register_cpu_ci(est_time=5, suite="stage-b-test-cpu-intel")
 
 
 class DummyDetector(BaseFormatDetector):
diff --git a/test/registered/unit/layers/moe/test_aiter_runner.py b/test/registered/unit/layers/moe/test_aiter_runner.py
index 65b941a59..60faa2369 100644
--- a/test/registered/unit/layers/moe/test_aiter_runner.py
+++ b/test/registered/unit/layers/moe/test_aiter_runner.py
@@ -14,7 +14,7 @@ from sglang.srt.layers.moe.moe_runner.aiter import (
 from sglang.srt.layers.moe.moe_runner.base import MoeRunnerConfig
 from sglang.test.ci.ci_register import register_cpu_ci
 
-register_cpu_ci(est_time=7, suite="base-c-test-cpu")
+register_cpu_ci(est_time=7, suite="stage-b-test-cpu-intel")
 
 
 def _runner_input():
diff --git a/test/registered/unit/layers/moe/test_hpc_ops_runner_guard.py b/test/registered/unit/layers/moe/test_hpc_ops_runner_guard.py
index 6c9e441e2..e89b939da 100644
--- a/test/registered/unit/layers/moe/test_hpc_ops_runner_guard.py
+++ b/test/registered/unit/layers/moe/test_hpc_ops_runner_guard.py
@@ -11,7 +11,7 @@ from sglang.srt.layers.moe.utils import MoeA2ABackend, MoeRunnerBackend
 from sglang.srt.runtime_context import get_flags
 from sglang.test.ci.ci_register import register_cpu_ci
 
-register_cpu_ci(est_time=6, suite="base-c-test-cpu")
+register_cpu_ci(est_time=6, suite="stage-b-test-cpu-intel")
 
 
 @pytest.fixture
diff --git a/test/registered/unit/layers/moe/test_moe_runner_extensions.py b/test/registered/unit/layers/moe/test_moe_runner_extensions.py
index 028adddc0..0452dd8a1 100644
--- a/test/registered/unit/layers/moe/test_moe_runner_extensions.py
+++ b/test/registered/unit/layers/moe/test_moe_runner_extensions.py
@@ -32,7 +32,7 @@ from sglang.srt.lora.layers import FusedMoEWithLoRA
 from sglang.srt.runtime_context import get_context, get_flags, get_parallel
 from sglang.test.ci.ci_register import register_cpu_ci
 
-register_cpu_ci(est_time=15, suite="base-c-test-cpu")
+register_cpu_ci(est_time=15, suite="stage-b-test-cpu-intel")
 
 
 class _TestDispatchRunnerCore(DispatchMoeRunnerCore):
diff --git a/test/registered/unit/managers/test_io_struct.py b/test/registered/unit/managers/test_io_struct.py
index a44abaa8d..40c534f9d 100644
--- a/test/registered/unit/managers/test_io_struct.py
+++ b/test/registered/unit/managers/test_io_struct.py
@@ -39,7 +39,7 @@ from sglang.test.test_utils import (
 
 register_cuda_ci(est_time=8, stage="base-b", runner_config="1-gpu-large")
 register_amd_ci(est_time=8, suite="stage-b-test-1-gpu-small-amd")
-register_cpu_ci(est_time=6, suite="base-c-test-cpu")
+register_cpu_ci(est_time=6, suite="stage-b-test-cpu-intel")
 
 
 class TestTokenizedReqInputMsgpack(unittest.TestCase):
diff --git a/test/registered/unit/managers/test_mm_embed_scatter.py b/test/registered/unit/managers/test_mm_embed_scatter.py
index bd7810af2..1a2361e76 100644
--- a/test/registered/unit/managers/test_mm_embed_scatter.py
+++ b/test/registered/unit/managers/test_mm_embed_scatter.py
@@ -4,7 +4,7 @@ import torch
 from sglang.srt.managers.mm_utils import _scatter_mm_embedding
 from sglang.test.ci.ci_register import register_cpu_ci
 
-register_cpu_ci(est_time=5, suite="base-b-test-cpu")
+register_cpu_ci(est_time=5, suite="stage-a-test-cpu-intel")
 
 NUM_TOKENS = 64
 
diff --git a/test/registered/unit/managers/test_mm_embedding_length.py b/test/registered/unit/managers/test_mm_embedding_length.py
index 0a6823fd4..b803be328 100644
--- a/test/registered/unit/managers/test_mm_embedding_length.py
+++ b/test/registered/unit/managers/test_mm_embedding_length.py
@@ -7,7 +7,7 @@ from sglang.srt.environ import envs
 from sglang.srt.managers import mm_schedule as mm_utils
 from sglang.test.ci.ci_register import register_cpu_ci
 
-register_cpu_ci(est_time=8, suite="base-b-test-cpu")
+register_cpu_ci(est_time=8, suite="stage-a-test-cpu-intel")
 
 
 @pytest.mark.parametrize(
diff --git a/test/registered/unit/managers/test_msgpack_ipc_roundtrip.py b/test/registered/unit/managers/test_msgpack_ipc_roundtrip.py
index 378f22b2a..a206c27fb 100644
--- a/test/registered/unit/managers/test_msgpack_ipc_roundtrip.py
+++ b/test/registered/unit/managers/test_msgpack_ipc_roundtrip.py
@@ -42,7 +42,7 @@ from sglang.srt.utils.weight_versions import WeightVersionSpan
 from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.test_utils import CustomTestCase
 
-register_cpu_ci(est_time=8, suite="base-c-test-cpu")
+register_cpu_ci(est_time=8, suite="stage-b-test-cpu-intel")
 
 
 def _round_trip(obj):
diff --git a/test/registered/unit/managers/test_scheduler_flush_cache.py b/test/registered/unit/managers/test_scheduler_flush_cache.py
index d80868762..233a45357 100644
--- a/test/registered/unit/managers/test_scheduler_flush_cache.py
+++ b/test/registered/unit/managers/test_scheduler_flush_cache.py
@@ -13,7 +13,7 @@ from sglang.srt.managers.scheduler_components.flush_wrapper import (
 )
 
 register_cpu_ci(est_time=14, suite="base-a-test-cpu")
-register_cpu_ci(est_time=8, suite="base-c-test-cpu")
+register_cpu_ci(est_time=8, suite="stage-b-test-cpu-intel")
 
 
 class TestSchedulerFlushCache(unittest.TestCase):
diff --git a/test/registered/unit/managers/test_scheduler_pause_generation.py b/test/registered/unit/managers/test_scheduler_pause_generation.py
index f2711acc9..31e8650eb 100644
--- a/test/registered/unit/managers/test_scheduler_pause_generation.py
+++ b/test/registered/unit/managers/test_scheduler_pause_generation.py
@@ -24,7 +24,7 @@ from sglang.srt.runtime_context import publish, reset_context
 from sglang.srt.sampling.sampling_params import SamplingParams
 
 register_cpu_ci(est_time=15, suite="base-a-test-cpu")
-register_cpu_ci(est_time=8, suite="base-c-test-cpu")
+register_cpu_ci(est_time=8, suite="stage-b-test-cpu-intel")
 
 
 class TestSchedulerPauseGeneration(unittest.TestCase):
diff --git a/test/registered/unit/mem_cache/test_evict_policy.py b/test/registered/unit/mem_cache/test_evict_policy.py
index ea94793c0..1d1fbe069 100644
--- a/test/registered/unit/mem_cache/test_evict_policy.py
+++ b/test/registered/unit/mem_cache/test_evict_policy.py
@@ -3,7 +3,7 @@
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(est_time=6, suite="base-a-test-cpu")
-register_cpu_ci(est_time=5, suite="base-c-test-cpu")
+register_cpu_ci(est_time=5, suite="stage-b-test-cpu-intel")
 
 import unittest
 from unittest.mock import MagicMock
diff --git a/test/registered/unit/observability/test_cpu_monitor.py b/test/registered/unit/observability/test_cpu_monitor.py
index 7027df69c..3d0e18a58 100644
--- a/test/registered/unit/observability/test_cpu_monitor.py
+++ b/test/registered/unit/observability/test_cpu_monitor.py
@@ -7,7 +7,7 @@ from unittest.mock import MagicMock, patch
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(est_time=60, suite="base-a-test-cpu", nightly=True)
-register_cpu_ci(est_time=5, suite="base-c-test-cpu")
+register_cpu_ci(est_time=5, suite="stage-b-test-cpu-intel")
 
 
 class TestCpuMonitor(unittest.TestCase):
diff --git a/test/registered/unit/observability/test_func_timer.py b/test/registered/unit/observability/test_func_timer.py
index fa50e5e8e..6f646928c 100644
--- a/test/registered/unit/observability/test_func_timer.py
+++ b/test/registered/unit/observability/test_func_timer.py
@@ -3,7 +3,7 @@
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(est_time=6, suite="base-a-test-cpu")
-register_cpu_ci(est_time=4, suite="base-c-test-cpu")
+register_cpu_ci(est_time=4, suite="stage-b-test-cpu-intel")
 
 import asyncio
 import unittest
diff --git a/test/registered/unit/observability/test_label_transform.py b/test/registered/unit/observability/test_label_transform.py
index 978d871c6..aa4857e03 100644
--- a/test/registered/unit/observability/test_label_transform.py
+++ b/test/registered/unit/observability/test_label_transform.py
@@ -3,7 +3,7 @@
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(est_time=6, suite="base-a-test-cpu")
-register_cpu_ci(est_time=4, suite="base-c-test-cpu")
+register_cpu_ci(est_time=4, suite="stage-b-test-cpu-intel")
 
 import unittest
 
diff --git a/test/registered/unit/observability/test_metrics_utils.py b/test/registered/unit/observability/test_metrics_utils.py
index 2be98978f..11ed06183 100644
--- a/test/registered/unit/observability/test_metrics_utils.py
+++ b/test/registered/unit/observability/test_metrics_utils.py
@@ -7,7 +7,7 @@ from sglang.srt.observability.utils import (
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(est_time=6, suite="base-a-test-cpu")
-register_cpu_ci(est_time=5, suite="base-c-test-cpu")
+register_cpu_ci(est_time=5, suite="stage-b-test-cpu-intel")
 
 
 class TestMetricsUtils(unittest.TestCase):
diff --git a/test/registered/unit/observability/test_request_metrics_exporter.py b/test/registered/unit/observability/test_request_metrics_exporter.py
index 21a99b42b..32f407c90 100644
--- a/test/registered/unit/observability/test_request_metrics_exporter.py
+++ b/test/registered/unit/observability/test_request_metrics_exporter.py
@@ -6,7 +6,7 @@ from typing import Any, Dict, List, Optional
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(est_time=6, suite="base-a-test-cpu")
-register_cpu_ci(est_time=5, suite="base-c-test-cpu")
+register_cpu_ci(est_time=5, suite="stage-b-test-cpu-intel")
 
 import asyncio
 import json
diff --git a/test/registered/unit/observability/test_startup_func_log_and_timer.py b/test/registered/unit/observability/test_startup_func_log_and_timer.py
index 67ab25121..e75497daf 100644
--- a/test/registered/unit/observability/test_startup_func_log_and_timer.py
+++ b/test/registered/unit/observability/test_startup_func_log_and_timer.py
@@ -3,7 +3,7 @@
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(est_time=6, suite="base-a-test-cpu")
-register_cpu_ci(est_time=4, suite="base-c-test-cpu")
+register_cpu_ci(est_time=4, suite="stage-b-test-cpu-intel")
 
 import unittest
 from unittest.mock import MagicMock, patch
diff --git a/test/registered/unit/parser/test_code_completion_parser.py b/test/registered/unit/parser/test_code_completion_parser.py
index 0acdb8e98..c62db905b 100644
--- a/test/registered/unit/parser/test_code_completion_parser.py
+++ b/test/registered/unit/parser/test_code_completion_parser.py
@@ -19,7 +19,7 @@ from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.test_utils import CustomTestCase
 
 register_cpu_ci(est_time=7, suite="base-a-test-cpu")
-register_cpu_ci(est_time=6, suite="base-c-test-cpu")
+register_cpu_ci(est_time=6, suite="stage-b-test-cpu-intel")
 
 
 class TestRegisterCompletionTemplate(CustomTestCase):
diff --git a/test/registered/unit/parser/test_conversation.py b/test/registered/unit/parser/test_conversation.py
index 7ac627b94..8f7f3b013 100644
--- a/test/registered/unit/parser/test_conversation.py
+++ b/test/registered/unit/parser/test_conversation.py
@@ -33,7 +33,7 @@ from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.test_utils import CustomTestCase
 
 register_cpu_ci(est_time=7, suite="base-a-test-cpu")
-register_cpu_ci(est_time=8, suite="base-c-test-cpu")
+register_cpu_ci(est_time=8, suite="stage-b-test-cpu-intel")
 
 
 class TestConversationGetPrompt(CustomTestCase):
diff --git a/test/registered/unit/parser/test_harmony_parser.py b/test/registered/unit/parser/test_harmony_parser.py
index 30f90eef6..7172eeca9 100644
--- a/test/registered/unit/parser/test_harmony_parser.py
+++ b/test/registered/unit/parser/test_harmony_parser.py
@@ -13,7 +13,7 @@ from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.test_utils import CustomTestCase
 
 register_cpu_ci(est_time=7, suite="base-a-test-cpu")
-register_cpu_ci(est_time=8, suite="base-c-test-cpu")
+register_cpu_ci(est_time=8, suite="stage-b-test-cpu-intel")
 
 
 class TestPrefixHold(CustomTestCase):
diff --git a/test/registered/unit/parser/test_jinja_template_utils.py b/test/registered/unit/parser/test_jinja_template_utils.py
index 0e08ba885..790bcd39a 100644
--- a/test/registered/unit/parser/test_jinja_template_utils.py
+++ b/test/registered/unit/parser/test_jinja_template_utils.py
@@ -12,7 +12,7 @@ from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.test_utils import CustomTestCase
 
 register_cpu_ci(est_time=7, suite="base-a-test-cpu")
-register_cpu_ci(est_time=6, suite="base-c-test-cpu")
+register_cpu_ci(est_time=6, suite="stage-b-test-cpu-intel")
 
 
 class TestTemplateContentFormatDetection(CustomTestCase):
diff --git a/test/registered/unit/parser/test_reasoning_content_without_parser.py b/test/registered/unit/parser/test_reasoning_content_without_parser.py
index dee92a07f..c2c3fc6e6 100644
--- a/test/registered/unit/parser/test_reasoning_content_without_parser.py
+++ b/test/registered/unit/parser/test_reasoning_content_without_parser.py
@@ -5,7 +5,7 @@ from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.test_utils import CustomTestCase
 
 register_cpu_ci(est_time=7, suite="base-a-test-cpu")
-register_cpu_ci(est_time=6, suite="base-c-test-cpu")
+register_cpu_ci(est_time=6, suite="stage-b-test-cpu-intel")
 
 # Simulated model output that contains think tags (e.g. from DeepSeek-R1)
 THINK_OUTPUT = (
diff --git a/test/registered/unit/sampling/test_custom_logit_processor.py b/test/registered/unit/sampling/test_custom_logit_processor.py
index 2e94311ec..e3c0e0a1b 100644
--- a/test/registered/unit/sampling/test_custom_logit_processor.py
+++ b/test/registered/unit/sampling/test_custom_logit_processor.py
@@ -3,7 +3,7 @@
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(est_time=7, suite="base-a-test-cpu")
-register_cpu_ci(est_time=8, suite="base-c-test-cpu")
+register_cpu_ci(est_time=8, suite="stage-b-test-cpu-intel")
 
 import json
 import unittest
diff --git a/test/registered/unit/sampling/test_penaltylib.py b/test/registered/unit/sampling/test_penaltylib.py
index 510c60d30..e25144d2e 100644
--- a/test/registered/unit/sampling/test_penaltylib.py
+++ b/test/registered/unit/sampling/test_penaltylib.py
@@ -3,7 +3,7 @@
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(est_time=9, suite="base-a-test-cpu")
-register_cpu_ci(est_time=8, suite="base-c-test-cpu")
+register_cpu_ci(est_time=8, suite="stage-b-test-cpu-intel")
 
 import unittest
 from unittest.mock import MagicMock
diff --git a/test/registered/unit/sampling/test_sampling_batch_info.py b/test/registered/unit/sampling/test_sampling_batch_info.py
index a5c3cef1e..475171a09 100644
--- a/test/registered/unit/sampling/test_sampling_batch_info.py
+++ b/test/registered/unit/sampling/test_sampling_batch_info.py
@@ -3,7 +3,7 @@
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(est_time=9, suite="base-a-test-cpu")
-register_cpu_ci(est_time=6, suite="base-c-test-cpu")
+register_cpu_ci(est_time=6, suite="stage-b-test-cpu-intel")
 
 import unittest
 from types import SimpleNamespace
diff --git a/test/registered/unit/sampling/test_sampling_params.py b/test/registered/unit/sampling/test_sampling_params.py
index cd1fe3cfc..d23ef24ed 100644
--- a/test/registered/unit/sampling/test_sampling_params.py
+++ b/test/registered/unit/sampling/test_sampling_params.py
@@ -3,7 +3,7 @@
 from sglang.test.ci.ci_register import register_cpu_ci, register_xpu_ci
 
 register_cpu_ci(est_time=7, suite="base-a-test-cpu")
-register_cpu_ci(est_time=8, suite="base-c-test-cpu")
+register_cpu_ci(est_time=8, suite="stage-b-test-cpu-intel")
 register_xpu_ci(est_time=10, suite="stage-a-test-1-gpu-xpu")
 
 import copy
diff --git a/test/registered/unit/server_args/test_server_args.py b/test/registered/unit/server_args/test_server_args.py
index b81bdd2df..009e312a6 100644
--- a/test/registered/unit/server_args/test_server_args.py
+++ b/test/registered/unit/server_args/test_server_args.py
@@ -94,7 +94,7 @@ from sglang.test.test_utils import (
 )
 
 register_cpu_ci(est_time=10, suite="base-a-test-cpu")
-register_cpu_ci(est_time=11, suite="base-c-test-cpu")
+register_cpu_ci(est_time=11, suite="stage-b-test-cpu-intel")
 
 # Mock get_device() so all tests run on CPU-only CI runners
 _mock_device = patch(
diff --git a/test/registered/unit/test_runai_utils.py b/test/registered/unit/test_runai_utils.py
index 244b1633d..f5a28129a 100644
--- a/test/registered/unit/test_runai_utils.py
+++ b/test/registered/unit/test_runai_utils.py
@@ -6,7 +6,7 @@ from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.test_utils import CustomTestCase
 
 register_cpu_ci(est_time=7, suite="base-a-test-cpu")
-register_cpu_ci(est_time=6, suite="base-c-test-cpu")
+register_cpu_ci(est_time=6, suite="stage-b-test-cpu-intel")
 
 
 class TestRunaiUtils(CustomTestCase):
diff --git a/test/registered/unit/utils/test_diffusion_torch_fallback.py b/test/registered/unit/utils/test_diffusion_torch_fallback.py
index 624c77978..c9a567b38 100644
--- a/test/registered/unit/utils/test_diffusion_torch_fallback.py
+++ b/test/registered/unit/utils/test_diffusion_torch_fallback.py
@@ -14,7 +14,7 @@ from sglang.kernels.ops.diffusion.common.fallback_torch import (
 from sglang.test.ci.ci_register import register_cpu_ci, register_mlx_ci
 
 register_cpu_ci(est_time=1, suite="base-a-test-cpu")
-register_cpu_ci(est_time=1, suite="base-b-test-cpu")
+register_cpu_ci(est_time=1, suite="stage-a-test-cpu-intel")
 register_cpu_ci(est_time=1, suite="base-b-test-cpu-arm64")
 register_mlx_ci(est_time=1, suite="stage-a-unit-test-mlx")
 
diff --git a/test/registered/unit/utils/test_gauge_histogram.py b/test/registered/unit/utils/test_gauge_histogram.py
index f78a2a3f3..438b34973 100644
--- a/test/registered/unit/utils/test_gauge_histogram.py
+++ b/test/registered/unit/utils/test_gauge_histogram.py
@@ -1,7 +1,7 @@
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(est_time=6, suite="base-a-test-cpu")
-register_cpu_ci(est_time=4, suite="base-c-test-cpu")
+register_cpu_ci(est_time=4, suite="stage-b-test-cpu-intel")
 
 import unittest
 
diff --git a/test/registered/unit/utils/test_http_server_auth.py b/test/registered/unit/utils/test_http_server_auth.py
index 690fa3e6f..87e1f400e 100644
--- a/test/registered/unit/utils/test_http_server_auth.py
+++ b/test/registered/unit/utils/test_http_server_auth.py
@@ -11,7 +11,7 @@ from sglang.srt.utils.auth import AuthLevel, decide_request_auth
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(est_time=6, suite="base-a-test-cpu")
-register_cpu_ci(est_time=5, suite="base-c-test-cpu")
+register_cpu_ci(est_time=5, suite="stage-b-test-cpu-intel")
 
 
 class TestHttpServerAdminAuth(unittest.TestCase):
diff --git a/test/registered/unit/utils/test_json_response.py b/test/registered/unit/utils/test_json_response.py
index ae3fa72b4..55c310945 100644
--- a/test/registered/unit/utils/test_json_response.py
+++ b/test/registered/unit/utils/test_json_response.py
@@ -11,7 +11,7 @@ from sglang.srt.utils.json_response import (
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(est_time=6, suite="base-a-test-cpu")
-register_cpu_ci(est_time=5, suite="base-c-test-cpu")
+register_cpu_ci(est_time=5, suite="stage-b-test-cpu-intel")
 
 
 class TestJSONResponseUtils(unittest.TestCase):
diff --git a/test/registered/unit/utils/test_patch_tokenizer.py b/test/registered/unit/utils/test_patch_tokenizer.py
index 17a9318ce..f5e912d5a 100644
--- a/test/registered/unit/utils/test_patch_tokenizer.py
+++ b/test/registered/unit/utils/test_patch_tokenizer.py
@@ -12,7 +12,7 @@ from sglang.srt.utils.patch_tokenizer import (
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(est_time=30, suite="base-a-test-cpu", nightly=True)
-register_cpu_ci(est_time=16, suite="base-c-test-cpu")
+register_cpu_ci(est_time=16, suite="stage-b-test-cpu-intel")
 
 
 class TestPatchTokenizerEndToEndTest(unittest.TestCase):
diff --git a/test/registered/unit/utils/test_subprocess_watchdog.py b/test/registered/unit/utils/test_subprocess_watchdog.py
index 6d62a65f0..a893c1c75 100644
--- a/test/registered/unit/utils/test_subprocess_watchdog.py
+++ b/test/registered/unit/utils/test_subprocess_watchdog.py
@@ -25,7 +25,7 @@ from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.test_utils import CustomTestCase
 
 register_cpu_ci(est_time=9, suite="base-a-test-cpu")
-register_cpu_ci(est_time=8, suite="base-c-test-cpu")
+register_cpu_ci(est_time=8, suite="stage-b-test-cpu-intel")
 
 
 def healthy_worker():
diff --git a/test/registered/utils/test_log_utils.py b/test/registered/utils/test_log_utils.py
index 422dba29a..fab48d150 100644
--- a/test/registered/utils/test_log_utils.py
+++ b/test/registered/utils/test_log_utils.py
@@ -11,7 +11,7 @@ from sglang.srt.utils.log_utils import create_log_targets, log_json
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(est_time=6, suite="base-a-test-cpu")
-register_cpu_ci(est_time=4, suite="base-c-test-cpu")
+register_cpu_ci(est_time=4, suite="stage-b-test-cpu-intel")
 
 _LOG_PREFIX_RE = re.compile(r"^\[\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2}\] ")
 
diff --git a/test/registered/utils/test_network_address.py b/test/registered/utils/test_network_address.py
index 6a9a9d773..d3ca555b0 100644
--- a/test/registered/utils/test_network_address.py
+++ b/test/registered/utils/test_network_address.py
@@ -7,7 +7,7 @@ from sglang.srt.utils.network import NetworkAddress, is_zmq_endpoint_ipv6
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(est_time=7, suite="base-a-test-cpu")
-register_cpu_ci(est_time=5, suite="base-c-test-cpu")
+register_cpu_ci(est_time=5, suite="stage-b-test-cpu-intel")
 
 # Mock get_device() so ServerArgs tests run on CPU-only CI runners
 _mock_device = patch(
diff --git a/test/registered/utils/test_socket_utils.py b/test/registered/utils/test_socket_utils.py
index 00adac911..7ba20ad18 100644
--- a/test/registered/utils/test_socket_utils.py
+++ b/test/registered/utils/test_socket_utils.py
@@ -16,7 +16,7 @@ from sglang.test.test_utils import CustomTestCase
 from sglang.utils import normalize_base_url, release_port, reserve_port
 
 register_cpu_ci(est_time=7, suite="base-a-test-cpu")
-register_cpu_ci(est_time=8, suite="base-c-test-cpu")
+register_cpu_ci(est_time=8, suite="stage-b-test-cpu-intel")
 
 
 class TestTryBindSocket(CustomTestCase):
diff --git a/test/registered/utils/test_type_based_dispatcher.py b/test/registered/utils/test_type_based_dispatcher.py
index b9f9280a9..4f875d90e 100644
--- a/test/registered/utils/test_type_based_dispatcher.py
+++ b/test/registered/utils/test_type_based_dispatcher.py
@@ -12,7 +12,7 @@ from sglang.test.ci.ci_register import register_amd_ci, register_cpu_ci
 from sglang.utils import TypeBasedDispatcher
 
 register_amd_ci(est_time=10, suite="stage-b-test-1-gpu-small-amd")
-register_cpu_ci(est_time=6, suite="base-c-test-cpu")
+register_cpu_ci(est_time=6, suite="stage-b-test-cpu-intel")
 
 
 class TestTypeBasedDispatcher(unittest.TestCase):
diff --git a/test/registered/vlm/test_token_id_retokenize_e2e.py b/test/registered/vlm/test_token_id_retokenize_e2e.py
index 402ac513b..8fa796927 100644
--- a/test/registered/vlm/test_token_id_retokenize_e2e.py
+++ b/test/registered/vlm/test_token_id_retokenize_e2e.py
@@ -36,7 +36,7 @@ from sglang.test.test_utils import (
 )
 
 register_cuda_ci(est_time=300, stage="base-b", runner_config="1-gpu-large")
-register_cpu_ci(est_time=123, suite="base-c-test-cpu")
+register_cpu_ci(est_time=123, suite="stage-b-test-cpu-intel")
 
 
 def _data_uri():
diff --git a/test/registered/vlm/test_video_utils.py b/test/registered/vlm/test_video_utils.py
index d0e9b18f4..1ad3204f9 100644
--- a/test/registered/vlm/test_video_utils.py
+++ b/test/registered/vlm/test_video_utils.py
@@ -6,7 +6,7 @@ from sglang.srt.utils import sample_video_frames
 from sglang.test.ci.ci_register import register_cpu_ci
 
 register_cpu_ci(est_time=7, suite="base-a-test-cpu")
-register_cpu_ci(est_time=5, suite="base-c-test-cpu")
+register_cpu_ci(est_time=5, suite="stage-b-test-cpu-intel")
 
 
 class DummyVideo:
diff --git a/test/run_suite.py b/test/run_suite.py
index 02ed0126e..9f6303fc0 100644
--- a/test/run_suite.py
+++ b/test/run_suite.py
@@ -32,9 +32,9 @@ HW_MAPPING = {
 PER_COMMIT_SUITES = {
     HWBackend.CPU: [
         "base-a-test-cpu",
-        "base-b-test-cpu",
-        "base-b-tp-test-cpu",
-        "base-c-test-cpu",
+        "stage-a-test-cpu-intel",
+        "stage-a-tp-test-cpu-intel",
+        "stage-b-test-cpu-intel",
         "base-b-test-cpu-arm64",
     ],
     HWBackend.AMD: [
@@ -425,7 +425,7 @@ def main():
         required=True,
         help=(
             "Test suite to run. Accepts a comma-separated list of suites "
-            "(e.g. 'base-b-test-cpu,base-c-test-cpu'); their tests are unioned "
+            "(e.g. 'stage-a-test-cpu-intel,stage-b-test-cpu-intel'); their tests are unioned "
             "into one pool before partitioning."
         ),
     )