Skip to content

#recipebot [Ironwood][SGLang-TorchTPU] Qwen3-Coder-480B-A35B-Instruct-FP8 single-host recipe - #377

Open
syeda-anjum wants to merge 1 commit into
AI-Hypercomputer:mainfrom
syeda-anjum:sglang-qwen3-coder-480b-v7x-single-host
Open

syeda-anjum wants to merge 1 commit into
AI-Hypercomputer:mainfrom
syeda-anjum:sglang-qwen3-coder-480b-v7x-single-host

Conversation

@syeda-anjum

Copy link
Copy Markdown
Contributor

Summary

Adds a single-host SGLang-TorchTPU inference recipe for Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8 on Ironwood (TPU7x, 2x2x1, 4 chips / 8 TensorCores, --tp-size 8 --ep-size 8).

  • Public image: lmsysorg/sglang:v0.1.0-tpu (sglang-torchtpu release tpu-v0.1.0).
  • Layout follows the existing single-host recipes: README.md + qwen3-coder-480b-server.yaml, benchmark Pods inlined in the README (apply → logs → delete).
  • Server flags follow the upstream sglang-torchtpu benchmark/tpu/recipe/qwen3_coder_480b recipe (fp8 weights, fp8_e4m3 KV cache, pallas_tpu_rpa_batched attention, torch.compile).
  • Benchmark numbers are left as xxx placeholders per recipe policy.

Files

  • inference/ironwood/SGLang/SGLang-TorchTPU/Qwen3-Coder-480B/single-host/README.md
  • inference/ironwood/SGLang/SGLang-TorchTPU/Qwen3-Coder-480B/single-host/qwen3-coder-480b-server.yaml

Validation (uBench run_ids)

  • 1k/1k: sglang_inference-torchtpu_sglang/qwen3_coder_480b_fp8_v7x_single_host_1k_1k-2026-10-20_223036-7905e6dc-b83e-405d-b9c1-4e1cf4698b79
  • 8k/1k: sglang_inference-torchtpu_sglang/qwen3_coder_480b_fp8_v7x_single_host_8k_1k-2026-10-20_224014-beeac8c9-1067-42a2-b80e-9dee25c85a95

@syeda-anjum

Copy link
Copy Markdown
Contributor Author

@joeywan-google @anthonsu PTAL — SGLang-TorchTPU single-host recipe, same layout as the Gemma4-26B / vLLM-TorchTPU recipes. uBench run_ids are in the description.

@gemini-code-assist gemini-code-assist Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Code Review

This pull request introduces a single-host deployment guide and Kubernetes manifests for serving the Qwen3-Coder-480B-A35B-Instruct-FP8 model using SGLang (TorchTPU) on Ironwood TPUs. The feedback points out an incorrect directory path in the README's cd command, which incorrectly includes the repository root name and would cause an error when run from the root.

Apply the SGLang server manifest using [qwen3-coder-480b-server.yaml](./qwen3-coder-480b-server.yaml):

```shell
cd tpu-recipes/inference/ironwood/SGLang/SGLang-TorchTPU/Qwen3-Coder-480B/single-host

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

The command uses tpu-recipes/ as a prefix in the path. However, when users clone the repository, they typically run commands from the repository root directory (which is already tpu-recipes). Running this command from the repository root will fail with a "No such file or directory" error. It is better to use the relative path from the repository root.

Suggested change
cd tpu-recipes/inference/ironwood/SGLang/SGLang-TorchTPU/Qwen3-Coder-480B/single-host
cd inference/ironwood/SGLang/SGLang-TorchTPU/Qwen3-Coder-480B/single-host

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant