ProgramBench
Program construction tasks
codingactive
Score in %
Definition
- Version
- not in the notes
- Category
- coding
- Maintainer
- not in the notes
- Status
- active
- Home page
- not in the notes
Sources
- Claude Fable 5.1 Benchmarks, Cost and Capabilities | Vals AI vals.ai https://vals.ai/models/anthropic_claude-fable-5-1
- huggingface.co/moonshotai/Kimi-K3 https://huggingface.co/moonshotai/Kimi-K3
- huggingface.co/moonshotai/Kimi-K2.7-Code https://huggingface.co/moonshotai/Kimi-K2.7-Code
Scores at a Glance
Score in %
Higher is better for this unit.
1
Kimi K3Moonshot AIKimi Code harnessKimi K2.7 CodeMoonshot AIKimi Code CLI, thinking
Claude Fable 5.1AnthropicVals.ai, max effort
Kimi K3Moonshot AIKimi Code harness77.8 %
vendor-reported2Kimi K2.7 CodeMoonshot AIKimi Code CLI, thinking53.6 %
vendor-reported3Claude Fable 5.1AnthropicVals.ai, max effort7 %
independent0255075100
Where this comes from
Each bar is one model, with its most independent recorded score. Hatched bars are vendor-reported; the word beside each bar says who ran it. Settings and dates differ between rows; the Scores table below has every result.
Scores
| Model | Score | Who ran it | Setting | View or population | Date | Source |
|---|---|---|---|---|---|---|
| Kimi K3 | 77.8 % | vendor-reported | Kimi Code harness | – | – | huggingface.co |
| Kimi K2.7 Code | 53.6 % | vendor-reported | Kimi Code CLI, thinking | – | – | huggingface.co |
 MiMo-V2.6-Flash | 26 % | vendor-reported | – | MiMo-V2.6 model card comparison table | 21 Sep 2026 | huggingface.co |
| Claude Fable 5.1 | 7 % ±1.81 | independent | Vals.ai, max effort | – | – | vals.ai |
Where this comes from
Scores from different settings, harnesses or dates are not directly comparable; the row says which. Sorted by score, highest first, as a reading aid, not a ranking.
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "ProgramBench", https://aurion.semai.id/benchmarks/programbench/, records of October 2026.