Purpose

Corbanu index runs treat the scoring model as a deterministic function: the same request bytes must yield the same response bytes on any admitted host. Every published index artifact carries a replay proof in which two independent H200 machines scored the full 1,000-company universe and every raw response matched byte for byte (1,000 of 1,000 comparisons byte-identical in the published live-sample artifact). This runbook records the exact profile that achieves that, and the hazards that break it.

Locked model and runtime pin

Every element below is pinned by identifier or digest. A run that deviates on any row is rejected by the operator before scoring starts.

ElementLocked value
ModelQwen/Qwen3.8-27B-FP8
Model revision017b9c7af6b5689d5dd426a76e0bc077eb5ca20a
Runtime imagelmsysorg/sglang:nightly-dev-cu13-20260817-d91c3682@sha256:fa8774dd128600a09fd6d46670b06fb69a55dac8a3881e50ccf0916a45eb39af
HardwareNVIDIA H200, single GPU
Tensor parallelism1
Attention backendstriton (attention and linear attention)
Radix cacheoff
CUDA graphsoff (prefill and decode)
Overlap scheduleoff
Deterministic inferenceon
Random seed438916795
Context length32,768
Max running requests32
Execution profile idqwen3.8-27b-fp8-h200-sglang-strict-fixed32-schema-v2

Server launch

The host bootstrap downloads the pinned revision through huggingface_hub.snapshot_download, records GPU identity, driver, and Python version into the run directory, then launches SGLang bound to localhost:

python3 -m sglang.launch_server \
  --model-path "$model_path" \
  --served-model-name Qwen/Qwen3.8-27B-FP8 \
  --host 127.0.0.1 \
  --port 8000 \
  --trust-remote-code \
  --tp 1 \
  --context-length 32768 \
  --mem-fraction-static 0.75 \
  --chunked-prefill-size 4096 \
  --max-running-requests 32 \
  --reasoning-parser qwen3 \
  --enable-deterministic-inference \
  --disable-radix-cache \
  --random-seed 438916795 \
  --enable-metrics

The server is reachable only through a narrow HTTPS tunnel; it is bound to localhost and exposed to a single local operator process.

Request discipline

  • temperature: 0, top_p: 1, thinking disabled per request.
  • Companies are scored in fixed 32-request batches per host. The batch schedule is part of the profile: floating-point reduction order inside a batch depends on which requests share it, so the batch composition itself must replay.
  • The mandate compiler and every company score use strict JSON response schemas; a response that fails to parse fails the run rather than being repaired.
  • Requests are content-addressed: each request’s SHA-256 is recorded and must match across hosts before response bytes are even compared.

Two-host replay gate

A run executes on two independent H200 machines, primary and replay:

  1. The compiled mandate rubric is generated on both hosts concurrently; the run aborts unless the raw rubric bytes are identical.
  2. All 1,000 companies are scored on both hosts in the same fixed batch order. Both hosts run to completion so a failure on one never alters the other’s batch schedule.
  3. For every company, request_sha256 must match, then the raw response bytes must match exactly.
  4. Any batch-sensitive mismatch triggers strict replay recovery: the exact request is recomputed alone on otherwise idle machines on both hosts, and only a byte-identical strict pair is admitted.
  5. The published artifact carries the per-company primary and replay content hashes plus an aggregate comparison SHA-256, so the replay claim can be inspected company by company from the artifact alone.

Published evidence: the live-sample artifact at /benchmarks/agentic-index-live-samples-20260827.json embeds the full model_profile and a replay_proof block with byte_identical_count: 1000 of comparison_count: 1000, alongside a representative raw replay with matching content hashes.

Determinism hazards

Each of these produced or would produce drift, and each is closed by the profile above.

HazardWhy it breaks byte-level replayMitigation
Radix / prefix cacheCache hits change kernel execution paths between otherwise identical requests--disable-radix-cache
Dynamic batchingReduction order varies with co-scheduled requestsFixed 32-request batches; batch schedule locked in the profile
CUDA graphs and overlap schedulingCapture and replay reorder work relative to eager executionBoth disabled
Tensor parallelismCross-GPU reductions have no fixed summation orderSingle GPU, --tp 1
Model or tokenizer driftA silently updated repo changes weights or prompt bytesRevision pin 017b9c7a… verified at load
Runtime driftKernel changes across SGLang builds change numericsImage pinned by sha256 digest
SamplingAny stochastic decode path defeats replayGreedy decode: temperature 0, top_p 1, fixed seed, thinking off
Hardware class driftDifferent GPU generations produce different numericsAdmitted hardware locked to H200

Profile history

  • qwen3.8-27b-fp8-h200-sglang-deterministic-noradix-schema-v1: first deterministic profile; established the pinned runtime, greedy decode, and radix-cache-off baseline used for the published live samples.
  • qwen3.8-27b-fp8-h200-sglang-strict-fixed32-schema-v2: current admitted profile; adds the fixed 32-request batch contract (request_batch_size_per_host: 32), strict replay recovery, and the batch-byte-identical count in the published proof. The public index contract test rejects artifacts from superseded profiles.

Scope

This profile makes one pinned model, one pinned runtime, and one hardware class replayable at the byte level. It is a serving discipline, and it is sufficient for two-host replay gating of index runs. Cross-vendor or cross-generation portability requires the separate consensus-executor line of work and is out of scope here.