Benchmark
Vals Terminal-Bench 2.0
Agentic benchmarks measure how well models carry out tasks that take several steps. Scores are percentages.
- Models measured
- 63
- Scores recorded
- 63
- Sources
- 1
- IQ weight
- 2.04
Score distribution
- 39.8%
- Mean
- 40.4%
- Median
- 16.9%
- Std dev
- 18%
- 10th percentile
- 59.6%
- 90th percentile
- 2.2%
- Min
This benchmark has a weight of 2.04 for chat models, measured over 63 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 51 | Grok 4.1 Fast· Vals AI | 24.7% |
| 52 | Qwen3 Max· Vals AI | 24.7% |
| 53 | Qwen3.5 Flash· Vals AI | 24.7% |
| 54 | DeepSeek V3.1· Vals AI | 22.5% |
| 55 | gemini-2.5-flash-preview-09-2025thinking· Vals AI | 21.3% |
| 56 | GPT OSS 120B· Vals AI | 19.1% |
| 57 | Trinity Large Thinking· Vals AI | 18% |
| 58 | Mistral Small2603· Vals AI | 16.9% |
| 59 | GPT 4.1· Vals AI | 14.6% |
| 60 | Magistral Medium· Vals AI | 13.5% |
| 61 | Mistral Large 3· Vals AI | 9% |
| 62 | Command A 03 2025· Vals AI | 2.2% |
| 63 | Llama 4 Maverick· Vals AI | 2.2% |
Showing 51–63 of 63 results.