Published reference2026-07-31
SWE-bench Verified
swe-bench-verified · 500 tasks. Compare only matched task sets and protocols.
DeepSeek V4 Flash 0731 · Published reference · DeepSeek · Reported evaluation
↗ Published resultSWE-bench Verified
DeepSeek-V4-Flash model card · Comparison across Modes ↗Tokens per second · see measurement basis
As recorded on this receipt
Run configuration & provenance +
- Context
- Not recorded
- Temperature
- 1
- Wall time
- Not recorded
- Weight file
- Provider-managed / not reported
- Revision
- Not reported
- Reasoning
- Not reported
- Agent version
- Not reported
- Output token limit
- Not reported
- Dataset revision
- Not reported
- Harness host
- Not reported
- Model host
- Published reference
- Input tokens
- Not reported
- Output tokens
- Not reported
- Cached input tokens
- Not reported
- Agent turn / step limit
- Not reported
- Concurrency
- Not reported
- Weight file size
- Not reported
Throughput measurement: Measurement method not separately reported. This should not be treated as standardized decode-only speed.
DeepSeek Harness · Think Max · SWE Verified (Resolved). Same card lists High 78.6 and Non-think 73.7. Prefer Max.
SWE Verified (Resolved) V4-Flash Max = 79.0 from the official DeepSeek-V4-Flash README comparison table. High 78.6 / Non-think 73.7 on the same card. n=500. Not a LocalsOnly desk run. https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash
Run ID: research-deepseek-v4-flash-0731-published-reference-swe-bench-verifiedTask explorer
Passed tasks first, then other outcomes grouped by recorded issue. Use the filters to inspect a particular outcome.
No task records in this receipt.
The headline or summary is available above. Individual outcomes have not been provided.
Explore test coverage →