Stage
Scored
12 of 12 stages
Settled in
44h 44m
committed Oct 21, 16:08 UTC
Round score
0.2241
0 is baseline speed
p99 latency
—
ceiling 2000 / 50 ms
Round 4: scored · 0.2241
12/12 stages
Intake & gates
6/6Patch hash recorded; awaiting fetch.
{
"hotkey": "5DAhRt3tXgjbR2TrchYMD8wsce1mu2Ae8m9Xrh7NdiphUs3R",
"reason": "Kimi K3 16k-128k long-context workload",
"source": "operator_carryover",
"operator": "arpan",
"commit_block": 9291101,
"evidence_ref": "https://github.com/Pareton-ai/pareton/pull/206",
"patch_fingerprint": "sha256:e6cbe23f18e7b3fd0d115cbf3a0ab1d387c0f804bac87c15dd5dce195e30bdda",
"source_campaign_id": "9c32cdc5-4235-4166-8d47-fcd026970cce",
"source_submission_id": "4847c5af-94ec-4a98-b6a0-d7e304dae6b2"
}A worker claimed the submission and started the pipeline.
{
"job_id": 287
}Patch artifact retrieved from object storage.
{
"size": 612597,
"patch_hash": "sha256:b13c15158f1cb473327c7d2e2d8021f4bb71c8d83a6e5c840d62e6d6c556b73a"
}Integrity and path constraints checked.
{
"identity": "ok",
"patch_hash": "sha256:b13c15158f1cb473327c7d2e2d8021f4bb71c8d83a6e5c840d62e6d6c556b73a"
}Patch applied onto the baseline checkout.
{
"ok": true
}Public API / surface compatibility held.
{
"files": [
"python/sglang/kernels/jit/csrc/elementwise/l2_prefetch.cuh",
"python/sglang/kernels/jit/csrc/kimi_k3/comm/dcp_a2a.cuh",
"python/sglang/kernels/jit/csrc/kimi_k3/comm/ls2_ar.cuh",
"python/sglang/kernels/jit/csrc/kimi_k3/comm/ls2_norm.cuh",
"python/sglang/kernels/jit/csrc/kimi_k3/comm/up_ag.cuh",
"python/sglang/kernels/jit/csrc/kimi_k3/kda/k3_kda_fold_lf.cuh",
"python/sglang/kernels/jit/csrc/moe/k3_moe_grid_trim.cuh",
"python/sglang/kernels/jit/csrc/moe/route_quant_fused.cuh",
"python/sglang/kernels/jit/csrc/moe/route_radix.cuh",
"python/sglang/kernels/ops/attention/fla/kda_replayssm_spec_decode.py",
"python/sglang/kernels/ops/attention/k3_draft_fold.py",
"python/sglang/kernels/ops/attention/k3_mla_pq.py",
"python/sglang/kernels/ops/attention/kda_decode_mtp.py",
"python/sglang/kernels/ops/communication/all_reduce_residual.py",
"python/sglang/kernels/ops/communication/k3_dcp_a2a.py",
"python/sglang/kernels/ops/communication/k3_ls2_ar.py",
"python/sglang/kernels/ops/communication/k3_ls2_norm.py",
"python/sglang/kernels/ops/communication/k3_up_ag.py",
"python/sglang/kernels/ops/elementwise/l2_prefetch.py",
"python/sglang/kernels/ops/gemm/cutedsl_bf16_gemm.py",
"python/sglang/kernels/ops/gemm/cutedsl_bf16_gemm_k3sk.py",
"python/sglang/kernels/ops/gemm/kimi_k3.py",
"python/sglang/kernels/ops/moe/k3_moe_gemm2x.py",
"python/sglang/kernels/ops/moe/k3_moe_gemm2x_swap.py",
"python/sglang/kernels/ops/moe/k3_moe_grid_trim.py",
"python/sglang/kernels/ops/moe/moe_route_quant_fused.py",
"python/sglang/kernels/ops/speculative/dspark/dspark_accept.py",
"python/sglang/kernels/ops/speculative/dspark/fused_accept.py",
"python/sglang/kernels/ops/speculative/dspark/fused_markov.py",
"python/sglang/srt/arg_groups/cuda_graph_hook.py",
"python/sglang/srt/distributed/device_communicators/configs/custom_all_reduce_v2.py",
"python/sglang/srt/distributed/device_communicators/custom_all_reduce_v2.py",
"python/sglang/srt/distributed/device_communicators/triton_symm_mem_ag.py",
"python/sglang/srt/layers/attention/cutedsl_mla_backend.py",
"python/sglang/srt/layers/attention/linear/kda_backend.py",
"python/sglang/srt/layers/attention/trtllm_mha_backend.py",
"python/sglang/srt/layers/communication/k3_ar_fusion.py",
"python/sglang/srt/layers/dcp/comm.py",
"python/sglang/srt/layers/moe/k3_front_split.py",
"python/sglang/srt/layers/moe/moe_runner/flashinfer_trtllm.py",
"python/sglang/srt/layers/moe/moe_runner/k3_moe_premeta.py",
"python/sglang/srt/layers/moe/route_quant_handoff.py",
"python/sglang/srt/model_executor/model_runner.py",
"python/sglang/srt/model_executor/runner/flashinfer_autotune.py",
"python/sglang/srt/models/deepseek_common/attention_forward_methods/forward_mla.py",
"python/sglang/srt/models/dspark.py",
"python/sglang/srt/models/kimi_k3.py",
"python/sglang/srt/speculative/dspark_components/adaptive_verify.py",
"python/sglang/srt/speculative/dspark_components/dspark_draft.py",
"python/sglang/srt/speculative/dspark_components/dspark_draft_sampler.py",
"python/sglang/srt/speculative/dspark_components/dspark_verify.py",
"python/sglang/srt/speculative/dspark_components/dspark_worker_v2.py",
"python/sglang/srt/speculative/dspark_components/exact_accept.py",
"python/sglang/srt/speculative/spec_tp_sync.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@sha256:5d24d4910d918080b55f6ddcf8dbcec3bdb63a218bd1a96c064b699dcc88c40a",
"image_digest": "sha256:5d24d4910d918080b55f6ddcf8dbcec3bdb63a218bd1a96c064b699dcc88c40a"
}Candidate engine image built and digest pinned.
{
"mock": false,
"build_log": "/var/log/pareton/builds/67996ff0-872f-43f1-b8ea-84f2b22b07a8/build.log",
"image_ref": "ghcr.io/pareton-ai/pareton-engine@sha256:5d24d4910d918080b55f6ddcf8dbcec3bdb63a218bd1a96c064b699dcc88c40a",
"image_tag": "ghcr.io/pareton-ai/pareton-engine:b13c15158f1cb473327c7d2e2d8021f4bb71c8d83a6e5c840d62e6d6c556b73a"
}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": 2,
"round_id": "2b888546-7297-4a31-9da7-9d9bfc9ecf73"
}The image ran and holds a round score.
{
"score": 0.33031274227088486,
"reason": null,
"ordinal": 2,
"round_id": "2b888546-7297-4a31-9da7-9d9bfc9ecf73"
}