Skip to content

ci(benchmarks): add Devstral 2 to H200 nightly benchmarks #1069

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
smfirmin wants to merge 2 commits into
base: main
Choose a base branch
from
+25 −1
Open

ci(benchmarks): add Devstral 2 to H200 nightly benchmarks #1069

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
6c67776
ci(benchmarks): add Devstral 2 to H200 nightly benchmarks
smfirmin Apr 8, 2026
34a818c
Update .github/workflows/nightly-benchmark.yml
smfirmin Apr 8, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions .github/workflows/nightly-benchmark.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -350,6 +350,7 @@ jobs:
- { id: meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8, slug: meta-llama-Llama-4-Maverick-17B-128E-Instruct-FP8, test_class: TestNightlyLlama4MaverickSingle }
# Pausing MiniMax M2 nightly benchmark
# - { id: minimaxai/minimax-m2, slug: minimaxai-minimax-m2, test_class: TestNightlyMinimaxM2Single }
- {id: mistralai/Devstral-2-123B-Instruct-2512, slug: mistralai-Devstral-2-123B-Instruct-2512, test_class: TestNightlyDevstral2Single}
variant:
- { id: sglang, runtime: sglang, grpc_only: "false", setup_vllm: false, setup_trtllm: false, extra_deps: "genai-bench" }
- { id: vllm, runtime: vllm, grpc_only: "false", setup_vllm: true, setup_trtllm: false, extra_deps: "genai-bench" }
Expand Down
16 changes: 15 additions & 1 deletion e2e_test/benchmarks/test_nightly_perf.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -102,6 +102,13 @@ def _run_nightly(setup_backend, genai_bench_runner, model_id, worker_count=1, **
("Qwen/Qwen3-30B-A3B", "Qwen30b", 4, ["http", "grpc"], {}),
("openai/gpt-oss-20b", "GptOss20b", 1, ["http", "grpc"], {}),
("minimaxai/minimax-m2", "MinimaxM2", 1, ["http", "grpc"], {}),
(
"mistralai/Devstral-2-123B-Instruct-2512",
"Devstral2",
1,
["http", "grpc"],
{},
),
(
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
"Llama4Maverick",
Expand Down Expand Up @@ -132,6 +139,10 @@ def _run_nightly(setup_backend, genai_bench_runner, model_id, worker_count=1, **
),
]

_SINGLE_ONLY_NIGHTLY_MODELS = {
"mistralai/Devstral-2-123B-Instruct-2512",
}


# ---------------------------------------------------------------------------
# Dynamic test class generation
Expand Down Expand Up @@ -163,7 +174,10 @@ def test_nightly_perf(self, setup_backend, genai_bench_runner):


for _model_id, _name, _multi_workers, _backends, _extra in _NIGHTLY_MODELS:
for _suffix, _count in [("Single", 1), ("Multi", _multi_workers)]:
_variants = [("Single", 1)]
if _model_id not in _SINGLE_ONLY_NIGHTLY_MODELS:
_variants.append(("Multi", _multi_workers))
for _suffix, _count in _variants:
_cls_name = f"TestNightly{_name}{_suffix}"
_cls = _make_test_class(_model_id, _count, _backends, _extra)
_cls.__name__ = _cls_name
Expand Down
9 changes: 9 additions & 0 deletions e2e_test/infra/model_specs.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -114,6 +114,15 @@ def _resolve_model_path(hf_path: str) -> str:
"sglang_args": ["--trust-remote-code"],
"vllm_args": ["--trust-remote-code"],
},
# Devstral 2 123B - Nightly benchmarks
"mistralai/Devstral-2-123B-Instruct-2512": {
"model": _resolve_model_path("mistralai/Devstral-2-123B-Instruct-2512"),
"tp": 4,
"features": ["chat", "streaming", "function_calling", "reasoning"],
"startup_timeout": 1200,
"sglang_args": ["--trust-remote-code"],
"vllm_args": ["--trust-remote-code"],
},
# Vision-language model for multimodal benchmarks (MMMU)
"Qwen/Qwen3-VL-8B-Instruct": {
"model": _resolve_model_path("Qwen/Qwen3-VL-8B-Instruct"),
Expand Down