tests/evals/gpt_oss/configs/gpt-oss-20b-rocm-quark-mxfp4-bf16-aiter.yaml

# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
model_name: amd/gpt-oss-20b-w-mxfp4-a-bf16
metric_threshold: 0.568
reasoning_effort: low
server_args: "--attention-backend ROCM_AITER_UNIFIED_ATTN --moe-backend aiter"
env:
  VLLM_ROCM_USE_AITER: "1"
[CI][ROCm] Add gpt-oss w4a8 in CI (#38292) Signed-off-by: Bowen Bao <bowenbao@amd.com> 2026-04-02 09:06:01 -07:00			`# SPDX-License-Identifier: Apache-2.0`
			`# SPDX-FileCopyrightText: Copyright contributors to the vLLM project`
[ROCm][Quantization][1/N] Refactor quark_moe w_mxfp4 w/ oracle (#38774) Signed-off-by: Bowen Bao <bowenbao@amd.com> Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com> 2026-04-02 20:29:57 -07:00			`model_name: amd/gpt-oss-20b-w-mxfp4-a-bf16`
[CI][ROCm] Add gpt-oss w4a8 in CI (#38292) Signed-off-by: Bowen Bao <bowenbao@amd.com> 2026-04-02 09:06:01 -07:00			`metric_threshold: 0.568`
			`reasoning_effort: low`
[ROCm][Quantization][1/N] Refactor quark_moe w_mxfp4 w/ oracle (#38774) Signed-off-by: Bowen Bao <bowenbao@amd.com> Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com> 2026-04-02 20:29:57 -07:00			`server_args: "--attention-backend ROCM_AITER_UNIFIED_ATTN --moe-backend aiter"`
[CI][ROCm] Add gpt-oss w4a8 in CI (#38292) Signed-off-by: Bowen Bao <bowenbao@amd.com> 2026-04-02 09:06:01 -07:00			`env:`
[ROCm][Quantization][1/N] Refactor quark_moe w_mxfp4 w/ oracle (#38774) Signed-off-by: Bowen Bao <bowenbao@amd.com> Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com> 2026-04-02 20:29:57 -07:00			`VLLM_ROCM_USE_AITER: "1"`