docs(cookbook): fix stale/pruned Docker image tags (#31508)
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
co-authored by
Claude Opus 4.8
parent
2c856abbe3
commit
ae3f62613a
@@ -28,7 +28,7 @@ SGLang from the main branch is required for Nemotron3-Super. You can install fro
|
|||||||
uv pip install 'git+https://github.com/sgl-project/sglang.git#subdirectory=python'
|
uv pip install 'git+https://github.com/sgl-project/sglang.git#subdirectory=python'
|
||||||
|
|
||||||
# Or use Docker
|
# Or use Docker
|
||||||
docker pull lmsysorg/sglang:nightly-dev-20260310-0fd9a57d
|
docker pull lmsysorg/sglang:latest
|
||||||
```
|
```
|
||||||
|
|
||||||
For the full Docker setup and other installation methods, please refer to the [official SGLang installation guide](../../../docs/get-started/install).
|
For the full Docker setup and other installation methods, please refer to the [official SGLang installation guide](../../../docs/get-started/install).
|
||||||
|
|||||||
@@ -19,14 +19,14 @@ Step-3.7-Flash is currently available in SGLang via Docker image install.
|
|||||||
### Docker (NVIDIA)
|
### Docker (NVIDIA)
|
||||||
```bash Command
|
```bash Command
|
||||||
# Pull the docker image
|
# Pull the docker image
|
||||||
docker pull lmsysorg/sglang:dev-step-3.7-flash
|
docker pull lmsysorg/sglang:latest
|
||||||
|
|
||||||
# Launch the container
|
# Launch the container
|
||||||
docker run -it --gpus all \
|
docker run -it --gpus all \
|
||||||
--shm-size=32g \
|
--shm-size=32g \
|
||||||
--ipc=host \
|
--ipc=host \
|
||||||
--network=host \
|
--network=host \
|
||||||
lmsysorg/sglang:dev-step-3.7-flash bash
|
lmsysorg/sglang:latest bash
|
||||||
```
|
```
|
||||||
|
|
||||||
## 3. Model Deployment
|
## 3. Model Deployment
|
||||||
|
|||||||
@@ -42,7 +42,7 @@ docker run -it --gpus all \
|
|||||||
--shm-size=32g \
|
--shm-size=32g \
|
||||||
--ipc=host \
|
--ipc=host \
|
||||||
--network=host \
|
--network=host \
|
||||||
lmsysorg/sglang:nightly-dev-20251215-4449c170 bash
|
lmsysorg/sglang:latest bash
|
||||||
|
|
||||||
# If you already have SGLang installed, uninstall the current SGLang version
|
# If you already have SGLang installed, uninstall the current SGLang version
|
||||||
pip uninstall sglang -y
|
pip uninstall sglang -y
|
||||||
|
|||||||
@@ -54,16 +54,7 @@ tag: NEW
|
|||||||
|
|
||||||
Refer to the [official SGLang installation guide](../../../docs/get-started/install).
|
Refer to the [official SGLang installation guide](../../../docs/get-started/install).
|
||||||
|
|
||||||
**Docker Images by Variant × Hardware:**
|
**Docker Image:** All variants (MiMo-V2.5 310B and MiMo-V2.5-Pro 1.02T) use `lmsysorg/sglang:latest`, which ships CUDA 13.0 and runs on both Hopper (H100 / H200) and Blackwell (B200 / GB300).
|
||||||
|
|
||||||
| Variant | Hardware | Docker Image |
|
|
||||||
| --- | --- | --- |
|
|
||||||
| **MiMo-V2.5 (310B)** | H100 / H200 (Hopper, CUDA 12.9) | `lmsysorg/sglang:nightly-dev-20260511-044bb88a` |
|
|
||||||
| **MiMo-V2.5 (310B)** | B200 / GB300 (Blackwell, CUDA 13.0) | `lmsysorg/sglang:nightly-dev-cu13-20260511-044bb88a` |
|
|
||||||
| **MiMo-V2.5-Pro (1.02T)** | H100 / H200 (Hopper, CUDA 12.9) | `lmsysorg/sglang:nightly-dev-20260511-044bb88a` |
|
|
||||||
| **MiMo-V2.5-Pro (1.02T)** | B200 / GB300 (Blackwell, CUDA 13.0) | `lmsysorg/sglang:nightly-dev-cu13-20260511-044bb88a` |
|
|
||||||
|
|
||||||
> Pull the image matching your GPU's CUDA driver. `lmsysorg/sglang:latest` will not load either checkpoint.
|
|
||||||
|
|
||||||
**TPU (sgl-jax):** MiMo-V2.5-Pro can also be served on TPU via the JAX-based [sgl-jax](https://github.com/sgl-project/sglang-jax) runtime. The container image and `pip install` steps are listed in [§3.3 TPU Deployment](#3-3-tpu-deployment-mimo-v2-5-pro-sgl-jax).
|
**TPU (sgl-jax):** MiMo-V2.5-Pro can also be served on TPU via the JAX-based [sgl-jax](https://github.com/sgl-project/sglang-jax) runtime. The container image and `pip install` steps are listed in [§3.3 TPU Deployment](#3-3-tpu-deployment-mimo-v2-5-pro-sgl-jax).
|
||||||
|
|
||||||
|
|||||||
Reference in New Issue
Block a user