Favur Evals
favur.dev

Run · circles · Jul 2026

Deepseek Flash v4

65.8 / 100 composite

all-DeepSeek · completed Jul 2, 2026

850 reqs · 1h 30m

Most concise source code — 115 lines.

Every run on Favur Evals is scored by the same deterministic engine, on the same Statements of Work. The benchmark is self-funded — no vendor sponsorship, credits, or grants.