Favur Evals
favur.dev

FAVUR EVALS

Multi-agent SDLC, evaluated.

We run the same statement of work through different LLM configurations, observing how each behaves inside Favur's agentic process. 76 runs so far across every Statement of Work, more coming.

Self-funded — no vendor sponsorship, credits, or grants.Last updated Jul 30, 2026

favur.dev — the pipeline behind these runs
scroll

Loading the interactive dashboard…

Every run, statically

The complete corpus without JavaScript — every run links to its permanent page.

All scored runs ranked by composite score
#RunSoWDateCompositeCoverageRubric
1Gemini Flash 3.5 + Gemini Flash LitecirclesJul 202681.886%v15
2Mimo v2.5circlesJul 202681.299%v15
3Muse Spark 1.1circlesJul 202680.898%v15
4Plus qwen3.7circlesJul 202679.999%v15
5Gemini Flash 3.5 + Gemini Flash LitecirclesJul 202679.686%v15
6Gemini Flash Lite 3.5circlesJul 202678.299%v15
7Plus qwen3.7circlesJul 202676.799%v15
8Mimo v2.5solarSystemJul 202675.999%v15
9Gemini Flash 3.5 + Gemini Flash LitecirclesJul 202674.699%v15
10Gemini Flash 3.6circlesJul 202673.299%v15
11Gemini Flash 3.6 + Gemini Flash LitecirclesJul 202672.199%v15
12Claude Sonnet 5circlesJul 202670.899%v15
13Deepseek Flash v4circlesJul 202670.198%v15
14Mimo Pro v2.5 + MimoarenaEvolutionJul 202667.876%v15
15Muse Spark 1.1solarSystemJul 202667.299%v15
16Gemini Flash Lite 3.1circlesJul 202665.761%v15
17Gemini Flash 3.5 + Gemini Flash LitecirclesJul 202665.699%v15
18Mimo v2.5circlesJul 202664.962%v15
19Grok 4.5circlesJul 202664.799%v15
20Mimo Pro v2.5circlesJul 202664.362%v15
21Plus qwen3.7circlesJul 202664.361%v15
22Mimo Pro v2.5 + MimocirclesJul 202664.399%v15
23Gemini Flash Lite 3.1circlesJul 202664.262%v15
24Deepseek Flash v4circlesJul 202664.063%v15
25Gemini Flash 3.6 + Gemini Flash LitesolarSystemJul 202663.899%v15
26Kimi k2.6circlesJul 202663.399%v15
27Grok 4.5circlesJul 202663.299%v15
28Mimo Pro v2.5 + MimosolarSystemJul 202663.299%v15
29Mimo Pro v2.5circlesJun 202663.057%v15
30Grok 4.5circlesJul 202662.399%v15
31Gemini Flash Lite 3.1solarSystemJul 202662.263%v15
32Glm 5.2circlesJul 202661.699%v15
33Gpt Terra 5.6circlesJul 202661.565%v15
34Kimi Code k2.7circlesJul 202661.163%v15
35Gemini Flash 3.6circlesJul 202661.199%v15
36Gpt Luna 5.6circlesJul 202661.162%v15
37Grok 4.5circlesJul 202661.063%v15
38Kimi Code k2.7circlesJul 202659.362%v15
39Step Flash 3.7circlesJul 202658.961%v15
40Claude Sonnet 5 + Gemini Flash LitesolarSystemJul 202658.399%v15
41Claude Sonnet 5circlesJul 202657.999%v15
42Gemini Flash 3.5 + Gemini Flash LitesolarSystemJul 202657.664%v15
43Mimo Pro v2.5solarSystemJun 202657.657%v15
44Step Flash 3.7solarSystemJul 202656.999%v15
45Deepseek Flash v4circlesJul 202656.862%v15
46Muse Spark 1.1the2048Jul 202656.897%v15
47Gemini Flash 3 (preview)circlesJun 202656.658%v15
48Gemini Flash 3 (preview)circlesJul 202656.563%v15
49Deepseek Flash v4circlesJul 202656.063%v15
50Deepseek Pro v4circlesJul 202655.963%v15
51Deepseek Flash v4the2048Jun 202655.658%v15
52Plus qwen3.7solarSystemJul 202654.799%v15
53Glm 5.2circlesJul 202654.562%v15
54Gemini Flash 3 (preview)circlesJul 202654.462%v15
55Deepseek Pro v4circlesJun 202653.957%v15
56Gemini Flash 3 (preview)circlesJul 202653.863%v15
57Gemini Pro 3.1 (preview)circlesJul 202653.461%v15
58Kimi k2.6circlesJun 202653.157%v15
59Gemini Flash 3.6 + Gemini Flash Litethe2048Jul 202652.697%v15
60Gemini Flash 3.5circlesJul 202652.663%v15
61Deepseek Pro v4circlesJul 202651.962%v15
62Gemini Pro 3.1 (preview)circlesJun 202651.758%v15
63InklingcirclesJul 202651.799%v15
64Gemini Flash Lite 3.1circlesJul 202651.099%v15
65Deepseek Flash v4solarSystemJun 202650.257%v15
66Kimi k2.6circlesJul 202649.962%v15
67InklingcirclesJul 202648.698%v15
68Claude Sonnet 5circlesJul 202647.863%v15
69Gemini Flash 3.5circlesJul 202647.462%v15
70Glm 5.2circlesJun 202646.957%v15
71Claude Sonnet 5circlesJul 202646.662%v15
72Claude Sonnet 4.6circlesJul 202644.462%v15
73Gemini Flash 3.5 + Gemini Flash LitesolarSystemJul 202642.662%v15
74Claude Sonnet 4.6circlesJun 202642.357%v15
75Gemini Flash 3 (preview)circlesJul 202639.162%v15
76Gemini Flash 3 (preview)solarSystemJun 202628.958%v15