Benchmark
Terminal-Bench Hard
Agentic benchmarks measure how well models carry out tasks that take several steps. Scores are percentages. Higher is better.
- Models measured
- 5
- Scores recorded
- 5
- Sources
- 2
- IQ weight
- No weight published
Score distribution
- 27.3%
- Mean
- 30.5%
- Median
- 7.5%
- Std dev
- 14%
- 10th percentile
- 36%
- 90th percentile
- 14%
- Min
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Gemma 4 31B ITGemma 4 Technical Report · Table 5 · thinking mode unless noted· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 36% |
| 2 | North Mini Code 1.0· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 31.1% |
| 3 | MiMo V2 FlashMiMo-V2-Flash final evaluation · harness unspecified· Reported by Xiaomi MiMo· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 30.5% |
| 4 | Command A Plus 05 2026· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 25% |
| 5 | Gemma 4 26B A4BGemma 4 Technical Report · Table 5 · thinking mode unless noted· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 14% |
Data credits: Data by LLM Stats · BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.