29 September 2026 · Agentic game · browser FPS
Whiteout Protocol · Sushi 1.0.5
Completed the Whiteout Protocol browser FPS challenge; the archived JavaScript and CSS match this run's build.
What was checked
- Production build passed; the JavaScript bundle was 625358 bytes.
- Production preview served successfully.
- A 60-second gameplay bot reached wave 3 and exercised combat, damage, death, and redeploy.
- The recoil pattern was byte-identical across two page loads.
- Final runtime verification reported zero console errors, warnings, or exceptions.
Configuration
The model, serving engine and agent used for this execution.
| Field | Value | Detail |
|---|---|---|
| Model · primary | Qwen 3.8 Flash Next · primary | EXL3 2.625 bpw |
| Runtime | Sushi 1.0.5 | inference engine and exact build |
| Harness | Pi coding agent | the agent or client that drove the run |
| Hardware | Apple M5 Pro · 64 GB unified memory | hardware recorded for this execution |
| Workload | Whiteout Protocol · revision 1 | Interactive long-horizon coding-agent challenge; not directly comparable with a fixed task suite. |
| Context window | 131,072 tokens | maximum context configured for the serving runtime |
| Repetitions | 1 | executions recorded for this workload; one means an indicative result, not an estimate |
Inference measurements
Engine-side rates for this run. Each figure states its own statistic; figures from different runs are not interchangeable.
| Metric | Value | Definition |
|---|---|---|
| Cache hit ratio | 95.17 % | server-side aggregate · n=1 |
| Decode · output-weighted | 38.66 tok/s | output-token weighted · n=254 |
| Prefill · median | 418.65 tok/s | median after first restored-cache request · Includes restored-cache processing; this is not a cold-prefill measurement. |
| MTP acceptance | 64.89 % | mean |
Agent session counters
Agent-wide usage totals. These include reasoning, tool calls and waiting, and are not model decode speed.
| Metric | Value | Definition |
|---|---|---|
| Session elapsed | 184.4 min | session start to final response · n=1 |
| Model requests | 254 requests | total · n=1 |
| Tool calls | 294 calls | total · n=1 |
| Context compactions | 3 compactions | total · n=1 |
Context and token counters
Token totals as reported by the harness or engine. Cached and cumulative figures are not additive with fresh input.
| Metric | Value | Definition |
|---|---|---|
| Maximum prompt context | 114,598 tokens | maximum · n=1 |
| Output tokens | 248,463 tokens | total · n=1 |
| Fresh input | 936,455 tokens | total · n=1 |
| Cached input | 18,454,158 tokens | total · n=1 |
Request records & provenance
Aggregate measurements only; no per-request rows are included. The original prompt, prompt fingerprint, session transcript, server logs, and local paths are excluded.
Run ID: 2026-09-29-qwen38-flash-next-sushi-whiteout-protocol
One execution, with checks transcribed from the retained record. A completed session is not a quality score. Measurement method · All runs