Stage
Scored
12 of 12 stages
Settled in
4h 28m
committed Sep 22, 03:28 UTC
Round score
0.618
0 is baseline speed
p99 latency
—
ceiling 2000 / 50 ms
Round 5: scored · 0.618
12/12 stages
Intake & gates
6/6Patch hash recorded; awaiting fetch.
{
"hotkey": "5Gy2ymrSM2CLiZhtM4oyfca4CBrmwq8Brbez4oAo8fcvB5Vc",
"payment_tx": 14,
"commit_block": 9120437,
"payment_block": 9120434,
"patch_fingerprint": "sha256:e661a29ae1be9d80a71e9ccab2ecb59da0247e7432e740a6bb7de5040b0fc07e",
"patch_reveal_delayed": true
}A worker claimed the submission and started the pipeline.
{
"job_id": 174
}Patch artifact retrieved from object storage.
{
"size": 387778,
"patch_hash": "sha256:c5ddfbc454b11091ddf7b0e212e8f9f4ec0c3cbd5c6aed99156a6111085b7eda"
}Integrity and path constraints checked.
{
"identity": "ok",
"patch_hash": "sha256:c5ddfbc454b11091ddf7b0e212e8f9f4ec0c3cbd5c6aed99156a6111085b7eda"
}Patch applied onto the baseline checkout.
{
"ok": true
}Public API / surface compatibility held.
{
"files": [
"python/sglang/kernels/ops/attention/fla/fused_sigmoid_gating_recurrent.py",
"python/sglang/kernels/ops/attention/fla/gdn_replayssm_deferred_fold.py",
"python/sglang/kernels/ops/attention/fla/gdn_replayssm_spec_decode.py",
"python/sglang/kernels/ops/attention/fla/layernorm_gated.py",
"python/sglang/kernels/ops/mamba/causal_conv1d_triton.py",
"python/sglang/kernels/ops/sampling/sharded_gumbel.py",
"python/sglang/kernels/ops/speculative/__init__.py",
"python/sglang/kernels/ops/speculative/topk1.py",
"python/sglang/srt/arg_groups/attention_hook.py",
"python/sglang/srt/arg_groups/speculative_hook.py",
"python/sglang/srt/distributed/device_communicators/fp8_a2a_all_reduce.py",
"python/sglang/srt/distributed/device_communicators/host_shm_ag_gemm.py",
"python/sglang/srt/distributed/device_communicators/host_shm_all_reduce.py",
"python/sglang/srt/distributed/parallel_state.py",
"python/sglang/srt/entrypoints/openai/serving_completions.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/linear/kernels/gdn_triton.py",
"python/sglang/srt/layers/attention/linear/sn10_extend_graph.py",
"python/sglang/srt/layers/attention/mamba/mamba2_metadata.py",
"python/sglang/srt/layers/communicator.py",
"python/sglang/srt/layers/logits_processor.py",
"python/sglang/srt/layers/quantization/fp8_utils.py",
"python/sglang/srt/layers/quantization/modelopt_quant.py",
"python/sglang/srt/layers/sampler.py",
"python/sglang/srt/layers/sharded_greedy.py",
"python/sglang/srt/layers/sharded_sampling.py",
"python/sglang/srt/layers/sn10_fp8_splitk.py",
"python/sglang/srt/layers/sn10_oproj_allgather.py",
"python/sglang/srt/managers/io_struct.py",
"python/sglang/srt/managers/schedule_batch.py",
"python/sglang/srt/managers/scheduler.py",
"python/sglang/srt/managers/scheduler_components/request_receiver.py",
"python/sglang/srt/managers/tokenizer_manager.py",
"python/sglang/srt/mem_cache/kv_cache_configurator.py",
"python/sglang/srt/mem_cache/mamba_slot_fused.py",
"python/sglang/srt/mem_cache/memory_pool.py",
"python/sglang/srt/mem_cache/unified_memory_pool.py",
"python/sglang/srt/model_executor/cpu_graph_runner.py",
"python/sglang/srt/model_executor/forward_batch_info.py",
"python/sglang/srt/model_executor/model_runner.py",
"python/sglang/srt/model_executor/runner/decode_cuda_graph_runner.py",
"python/sglang/srt/model_executor/runner/prefill_cuda_graph_runner.py",
"python/sglang/srt/models/qwen3_5.py",
"python/sglang/srt/speculative/eagle_draft_cuda_graph_runner.py",
"python/sglang/srt/speculative/eagle_draft_extend_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/sn10_fp8_draft_head.py",
"python/sglang/srt/speculative/spec_utils.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:b2779e0a405a25ecd594dcc693743e279ebe1deca188c4ed7201995116dfe407",
"image_digest": "sha256:b2779e0a405a25ecd594dcc693743e279ebe1deca188c4ed7201995116dfe407"
}Candidate engine image built and digest pinned.
{
"mock": false,
"build_log": "/var/log/pareton/builds/27995586-0160-433e-821a-7fa06cf7c992/build.log",
"image_ref": "ghcr.io/pareton-ai/pareton-engine@sha256:b2779e0a405a25ecd594dcc693743e279ebe1deca188c4ed7201995116dfe407",
"image_tag": "ghcr.io/pareton-ai/pareton-engine:c5ddfbc454b11091ddf7b0e212e8f9f4ec0c3cbd5c6aed99156a6111085b7eda"
}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": 4,
"round_id": "97544ca8-2fb9-4dc6-b1a7-b45bd0991667"
}The image ran and holds a round score.
{
"kind": "repeated-span-four-prompt-allowance-v1",
"score": 0.618003036260353,
"ordinal": 5,
"operator": "user-authorized Codex repair",
"round_id": "91ed633c-d5e4-4431-b523-7f682d8606bc",
"policy_pr": "https://github.com/Pareton-ai/pareton/pull/175",
"policy_commit": "4564159a31203e2f980f2555eeffbf73d4abd870",
"leader_decision": {
"event": null,
"prev_score": null,
"void_reason": null,
"leader_score": 0.6214243148870308,
"overtake_threshold": 0.6276385580359011,
"prev_submission_id": null,
"leader_submission_id": "4f455030-d1c7-4f53-aec2-45604b780bf2"
},
"previous_reason": "degenerate output (hf-023: rep 1: longest repeated span ratio 0.337 at or above harness ceiling over 20358 chars)",
"previous_status": "disqualified",
"source_evidence_sha256": {
"bench_report.json": "e1ec8c889b9260bf89ead61a4bd9d8efec2d4508c8502785074d8bce1ae81250",
"bench_request.remote.json": "01ea5608f688b105977b5abad498f2b6e811dcc66e90f06c80778c616c8be3a6",
"evidence/correctness/baseline.jsonl": "a95ce7cceeb0ffa2d34ee99c07490418b83966badd5169df36c13f688a5771de",
"evidence/correctness/candidate_2.jsonl": "1ce33414228e99d0930108501f13fb2f40a7785430fb8656593cd437231dc6ad",
"evidence/correctness/baseline_exclusions.json": "7773f6038e27aaa6c3d239beb29f2c9343ef25d5119360b264202fb7035184b4"
},
"repeated_span_degeneracy": {
"failed_prompts": 1,
"failed_request_ids": [
"hf-023"
],
"max_failed_prompts": 4
}
}Other events
The image ran and produced wrong output.
{
"score": null,
"reason": "degenerate output (hf-023: rep 1: longest repeated span ratio 0.337 at or above harness ceiling over 20358 chars)",
"ordinal": 5,
"round_id": "91ed633c-d5e4-4431-b523-7f682d8606bc"
}