Stage
Scored
12 of 12 stages
Settled in
2h 37m
committed Oct 12, 01:00 UTC
Round score
0.6581
0 is baseline speed
p99 latency
—
ceiling 2000 / 50 ms
Round 15: scored · 0.6581
12/12 stages
Intake & gates
6/6Patch hash recorded; awaiting fetch.
{
"hotkey": "5Fs6cf6fzdYF3VfiPBQacv1P5afbD4TRqHivoVzoiTFnJHyi",
"payment_tx": 11,
"commit_block": 9263680,
"payment_block": 9263677,
"patch_fingerprint": "sha256:626ef0d1e2a7459267156b9ef1a16f3a37bf419019936034635a2e415bfdeb8a"
}A worker claimed the submission and started the pipeline.
{
"job_id": 237
}Patch artifact retrieved from object storage.
{
"size": 400161,
"patch_hash": "sha256:4c2133081b1733ee70a43e70c211b84ff8bf0d9bf139b728e23a77cf02d4b285"
}Integrity and path constraints checked.
{
"identity": "ok",
"patch_hash": "sha256:4c2133081b1733ee70a43e70c211b84ff8bf0d9bf139b728e23a77cf02d4b285"
}Patch applied onto the baseline checkout.
{
"ok": true
}Public API / surface compatibility held.
{
"files": [
"python/sglang/kernels/jit/csrc/elementwise/lean_gemma_rmsnorm.cuh",
"python/sglang/kernels/jit/csrc/gemm/fp8_blockwise/fp8_blockwise_scaled_mm_sm120.cuh",
"python/sglang/kernels/jit/csrc/gemm/int4n/int4n_gemm.cuh",
"python/sglang/kernels/jit/csrc/gemm/marlin/gptq_marlin.cuh",
"python/sglang/kernels/jit/csrc/gemm/marlin/kernel.h",
"python/sglang/kernels/jit/csrc/gemm/marlin/marlin_template.h",
"python/sglang/kernels/ops/attention/fla/gdn_replayssm_spec_decode.py",
"python/sglang/kernels/ops/attention/fla/layernorm_gated.py",
"python/sglang/kernels/ops/attention/fused_qk_rmsnorm_rope_gate.py",
"python/sglang/kernels/ops/elementwise/cast_copy.py",
"python/sglang/kernels/ops/elementwise/elementwise.py",
"python/sglang/kernels/ops/embeddings/vocab_parallel_embedding.py",
"python/sglang/kernels/ops/gemm/fp8_blockwise_gemm.py",
"python/sglang/kernels/ops/kvcache/kv_indices.py",
"python/sglang/kernels/ops/layernorm/lean_norm.py",
"python/sglang/kernels/ops/mamba/causal_conv1d_triton.py",
"python/sglang/kernels/ops/quantization/gptq_marlin.py",
"python/sglang/kernels/ops/quantization/int4n_gemm.py",
"python/sglang/kernels/ops/speculative/attn_args.py",
"python/sglang/kernels/ops/speculative/cache_locs.py",
"python/sglang/kernels/ops/speculative/commit_indices.py",
"python/sglang/kernels/ops/speculative/reject_sampling.py",
"python/sglang/launch_server.py",
"python/sglang/srt/arg_groups/spec_xqa_defaults.py",
"python/sglang/srt/arg_groups/speculative_hook.py",
"python/sglang/srt/environ.py",
"python/sglang/srt/layers/attention/flashinfer_backend.py",
"python/sglang/srt/layers/attention/hybrid_linear_attn_backend.py",
"python/sglang/srt/layers/attention/linear/gdn_backend.py",
"python/sglang/srt/layers/attention/trtllm_mha_backend.py",
"python/sglang/srt/layers/l2_prefetch.py",
"python/sglang/srt/layers/layernorm.py",
"python/sglang/srt/layers/logits_processor.py",
"python/sglang/srt/layers/quantization/fp8.py",
"python/sglang/srt/layers/quantization/fp8_utils.py",
"python/sglang/srt/layers/quantization/marlin_a_prologue.py",
"python/sglang/srt/layers/quantization/nvfp4_decode.py",
"python/sglang/srt/layers/skinny_linear.py",
"python/sglang/srt/layers/vocab_parallel_embedding.py",
"python/sglang/srt/managers/overlap_utils.py",
"python/sglang/srt/managers/scheduler.py",
"python/sglang/srt/mem_cache/memory_pool.py",
"python/sglang/srt/model_executor/forward_batch_info.py",
"python/sglang/srt/model_executor/runner_backend/full_cuda_graph_backend.py",
"python/sglang/srt/models/qwen2_moe.py",
"python/sglang/srt/models/qwen3_5.py",
"python/sglang/srt/models/qwen3_5_mtp.py",
"python/sglang/srt/models/qwen3_vl.py",
"python/sglang/srt/speculative/draft_fp8_lm_head.py",
"python/sglang/srt/speculative/draft_vocab_subset.py",
"python/sglang/srt/speculative/eagle_draft_cuda_graph_runner.py",
"python/sglang/srt/speculative/eagle_info.py",
"python/sglang/srt/speculative/eagle_utils.py",
"python/sglang/srt/speculative/eagle_worker_common.py",
"python/sglang/srt/speculative/eagle_worker_v2.py",
"python/sglang/srt/speculative/fast_softmax.py",
"python/sglang/srt/speculative/spec_utils.py",
"python/sglang/srt/utils/host_marks.py"
]
}Build
3/3Hermetic engine image build in progress.
No structured detail recorded for this step.
Candidate engine image pushed to the registry.
{
"image_ref": "ghcr.io/pareton-ai/pareton-engine@sha256:87432bd62e3795255e9c4c8d5424feaa1d1d2cb4b345a87b3d9f2e0a75e6488a",
"image_digest": "sha256:87432bd62e3795255e9c4c8d5424feaa1d1d2cb4b345a87b3d9f2e0a75e6488a"
}Candidate engine image built and digest pinned.
{
"mock": false,
"build_log": "/var/log/pareton/builds/98265150-6024-411b-ac51-4977137c3e46/build.log",
"image_ref": "ghcr.io/pareton-ai/pareton-engine@sha256:87432bd62e3795255e9c4c8d5424feaa1d1d2cb4b345a87b3d9f2e0a75e6488a",
"image_tag": "ghcr.io/pareton-ai/pareton-engine:4c2133081b1733ee70a43e70c211b84ff8bf0d9bf139b728e23a77cf02d4b285"
}Round
3/3Waiting to be seated in a round.
No structured detail recorded for this step.
Seated in a round; waiting to run or running.
{
"ordinal": 15,
"round_id": "d8f31378-9bcc-4529-9701-1c88ad5bb168"
}The image ran and holds a round score.
{
"score": 0.6580820893035004,
"reason": null,
"ordinal": 15,
"round_id": "d8f31378-9bcc-4529-9701-1c88ad5bb168"
}