[Jiminator]

Motivation

The Beam Responses tool adapter in #45257 (responses_tools/beam.py) builds on ResponseFunctionShellToolCall, ShellCall and ShellCallOutput, which OpenAI 2.8.0 added. With 2.6.1 the server still starts, but creating the Responses handler raises ImportError. http_server.py then logs OpenAI Responses API (/v1/responses) disabled, and every /v1/responses request returns 500.

Split out of #44747 so the dependency can be reviewed on its own.

Modifications

Bumps openai from 2.6.1 to 2.8.0 in python/pyproject.toml, as #12222 did for 2.6.1.

The platform pyprojects (python/pyproject_{cpu,npu,other,xpu}.toml and 3rdparty/amd/wheel/sglang/pyproject.toml) stay at 2.6.1 for their owners to move. Their CI suites don't run the Beam Responses shell tests, and Beam's Responses path runs on CUDA.

Validation

• Beam on 4×GB300 and 8×H200 with OpenAI 2.6.1: chat completions and GSM8K are unaffected (95.68–96.89%, zero errors), and /v1/responses returns 500 with the logged ImportError.
• With 2.8.0, /v1/responses returns 200.
• CI on this PR installs python/pyproject.toml with dependency resolution, so the main suites run on 2.8.0.
• Not checked here: other models' Responses paths on 2.8.0 beyond what CI covers.

Merge order

Every part targets main. #44747 stays open, unchanged, as the full reference. Merge in this order:

1. [Chore] docker: Add Beam CUDA image (1/8) #45253 — Beam CUDA Docker image
2. [Chore] deps: Bump the OpenAI SDK to 2.8.0 for Responses shell tools (2/8) #45291 — OpenAI SDK 2.8.0, needed by the Responses shell tools in [Feat] parser: Add Beam reasoning, tool calls and response templates (5/7) #45257
3. [Feat] kernels: Beam model support #45254 — Kernels
4. [Feat] srt: Add attention, memory and DFlash DP support for Beam (4/8) #45255 — Attention, memory and DFlash DP support
5. [Feat] beam: Add Beam model, runtime support and checkpoints (4/7) #45256 — Beam model, checkpoints and tests
6. [Feat] parser: Add Beam reasoning, tool calls and response templates (5/7) #45257 — Beam reasoning, tool calls and response templates (also after [Feat] parser: Enrich chat_parsing streaming events #40478, [Fix] parser: Accept whole-number floats for integer tool arguments #43941 and [Feat] parser: Add checkpoint-driven response-template adapters #40479)
7. [Chore] deps: Add nvidia-cuda-tileiras for the optional cuTile router #45290 — nvidia-cuda-tileiras, needed by [Perf] beam: Add optional cuTile BF16x3 router GEMM #45361
8. [Perf] beam: Add optional cuTile BF16x3 router GEMM #45361 — cuTile BF16x3 router GEMM (optional: without it Beam uses the cuBLAS router)

#45292 (Dynamo and grpcio-tools pins for the Beam image) is not in the merge order for now.

🤖 Generated with Claude Code

---

CI States

Latest PR Test (Base): 🚫 Run #39502791400
Latest PR Test (Extra): ❌ Run #39502791330
Latest PR Test (AMD ROCm 10): ✅ Run #39502791399

[Jiminator]

/tag-and-rerun-ci

[Jiminator]

Passed all but 2 broken tests on old main: https://github.com/sgl-project/sglang/actions/runs/39372649730/job/118806942077

[Jiminator]

/rerun-test registered/kernels/ops/quantization/test_per_token_group_quant.py

[github-actions]

🚀 1-gpu-h100 (1 test): ✅ View workflow run

cd test/ && python3 registered/kernels/ops/quantization/test_per_token_group_quant.py

🚀 4-gpu-b200 (1 test): ✅ View workflow run

cd test/ && python3 registered/kernels/ops/quantization/test_per_token_group_quant.py