From 73d0989d9b5c2bd16191ac173f609c32db9535bb Mon Sep 17 00:00:00 2001 From: zijiexia <37504505+zijiexia@users.noreply.github.com> Date: Wed, 10 Jun 2026 14:04:16 -0700 Subject: [PATCH] docs: make playground issue template model field a free-form input (#27827) Co-authored-by: Claude Fable 5 --- .../cookbook-add-model/references/authoring-reference.md | 2 +- .../skills/cookbook-add-model/templates/config.jsx.tmpl | 5 +++-- .claude/skills/cookbook-review-pr/SKILL.md | 5 ++++- .github/ISSUE_TEMPLATE/3-playground-verified-cell.yml | 7 +++---- 4 files changed, 11 insertions(+), 8 deletions(-) diff --git a/.claude/skills/cookbook-add-model/references/authoring-reference.md b/.claude/skills/cookbook-add-model/references/authoring-reference.md index 9d35f2836..56d2177e2 100644 --- a/.claude/skills/cookbook-add-model/references/authoring-reference.md +++ b/.claude/skills/cookbook-add-model/references/authoring-reference.md @@ -48,7 +48,7 @@ the full contract): | `playgroundFeatures` | `{[axisId]: {...}}` | Opts into the Playground widget. See §2.3. | | `benchmarkCommands` | `{speed: string, accuracy: {[accKey]: string \| {[variant]: string}}, numPromptsByConc?: {[c]: number}}` | Powers the benchmark card's **"⚡ Reproduce"** modal. `speed` is ONE `bench_serving` template; the engine fills `{{DATASET}}`/`{{ISL}}`/`{{OSL}}` from each cell's `speed[].workload`, the chip-picked `{{MAX_CONCURRENCY}}`, and `{{NUM_PROMPTS}}` (resolved `workload.num_prompts ?? numPromptsByConc[c] ?? max(c*2, 200)`). `accuracy` maps an accuracy field (e.g. `gsm8k_pct`) to a per-eval template — a string, OR a `{flash, pro, …}` object keyed by variant when the command differs per variant (e.g. GPQA/AIME `--max-tokens`). The modal renders a chip per eval (one command area, like Speed). Both also use `{{MODEL_NAME}}` + `{{CURL_HOST}}`/`{{CURL_PORT}}` like `curl`. Optional; the button only appears when this AND `benchmarks` are present. | | `defaultAccuracy` | `{[variant]: {[accKey]: number}}` | Model-level accuracy applied to **every** cell of a variant (e.g. GPQA Diamond / AIME25 — hardware-independent). Merged UNDER each cell's measured `accuracy` (a per-cell value wins), so you set a variant's score once instead of copying it onto every benchmark entry. Keys must match `ACCURACY_LABELS` + `benchmarkCommands.accuracy`. | -| `github` | `{owner?, repo?, issueTemplate?, cookbookModel?}` | Overrides for the "Submit verified cell" CTA in the playground. Defaults: `sgl-project/sglang` + `3-playground-verified-cell.yml` + `"deepseek-ai/deepseek-v4"`. Set `cookbookModel` to the value that matches the `model` dropdown in your issue template so it's pre-selected when the issue opens. | +| `github` | `{owner?, repo?, issueTemplate?, cookbookModel?}` | Overrides for the "Submit verified cell" CTA in the playground. Defaults: `sgl-project/sglang` + `3-playground-verified-cell.yml` + `"deepseek-ai/deepseek-v4"`. Set `cookbookModel` to the model's HF id (`/`); it prefills the issue template's free-form `model` input when the issue opens. **Don't prune this block** — without it the engine falls back to `deepseek-ai/deepseek-v4` and submissions from your page get mislabeled. | ## 2.2 Author the 5-dim matrix (`cells[]`) diff --git a/.claude/skills/cookbook-add-model/templates/config.jsx.tmpl b/.claude/skills/cookbook-add-model/templates/config.jsx.tmpl index a5bec531a..0946237e9 100644 --- a/.claude/skills/cookbook-add-model/templates/config.jsx.tmpl +++ b/.claude/skills/cookbook-add-model/templates/config.jsx.tmpl @@ -142,8 +142,9 @@ sgl-eval run gsm8k \\ mi355x: "lmsysorg/sglang:dev-rocm720-mi35x", }, - // Pre-selects the issue template's `model` dropdown on "Submit verified cell". - // Must match that dropdown's value (usually `/`). + // Prefills the issue template's free-form `model` field on "Submit verified cell". + // Use the HF id (`/`). Do NOT delete this block when pruning — + // without it the engine falls back to "deepseek-ai/deepseek-v4" and mislabels submissions. github: { cookbookModel: "__HF_ORG__/__MODEL_SLUG__", }, diff --git a/.claude/skills/cookbook-review-pr/SKILL.md b/.claude/skills/cookbook-review-pr/SKILL.md index b4468cbec..1948db070 100644 --- a/.claude/skills/cookbook-review-pr/SKILL.md +++ b/.claude/skills/cookbook-review-pr/SKILL.md @@ -51,7 +51,10 @@ than restating. - `dockerImages` covers the hw ids that have cells (else users hit the `:dev` fallback). - `multiNodeHints` present ONLY for hw whose fabric needs manual NIC env (e.g. `gb200` NVL72) — NOT every `multi-N` hw (standard-IB DeepEP / Marlin multi-node don't need it). -- `github.cookbookModel` matches the issue-template `model` dropdown value. +- `github.cookbookModel` is set to the model's HF id (`/`). The issue + template's `model` field is a free-form input prefilled from this value; if the config + omits the `github` block, the engine falls back to `deepseek-ai/deepseek-v4` and the + page's submissions get mislabeled. - `playgroundFeatures` axes are pruned to what the model supports — no empty/stub axes (the `moe` axis's MegaMoE backend option + `megamoeQuant` block only on Blackwell MoE, gated by `requiresHw`; `hisparse` only DSA-style; `pdDisagg.router` only with a PD topology). diff --git a/.github/ISSUE_TEMPLATE/3-playground-verified-cell.yml b/.github/ISSUE_TEMPLATE/3-playground-verified-cell.yml index 88934b5d1..f48ea8211 100644 --- a/.github/ISSUE_TEMPLATE/3-playground-verified-cell.yml +++ b/.github/ISSUE_TEMPLATE/3-playground-verified-cell.yml @@ -21,13 +21,12 @@ body: recipe, they'll convert this issue into a PR against the cookbook cell catalog and close this issue. -- type: dropdown +- type: input id: model attributes: label: Cookbook model - description: Which model's cookbook does this cell belong to? - options: - - deepseek-ai/deepseek-v4 + description: Which model's cookbook does this cell belong to? Auto-filled by the playground — use the model id as it appears in the cookbook config. + placeholder: "deepseek-ai/deepseek-v4" validations: required: true