Benchmark
Finance Agent v2
Agentic benchmarks measure how well models carry out tasks that take several steps. Scores are percentages. Higher is better.
- Models measured
- 30
- Scores recorded
- 45
- Sources
- 2
- IQ weight
- 1.55
Score distribution
- 46.9%
- Mean
- 48.4%
- Median
- 8.9%
- Std dev
- 32.1%
- 10th percentile
- 57.9%
- 90th percentile
- 27.9%
- Min
This benchmark has a weight of 1.55 for chat models, measured over 30 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 |
|---|---|---|
| 1 | Gemini 3.8 Flashfinancial analyst tasks · pass@1· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 61.4% |
| 2 | Ling 3.0 Flash Fin· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 59.8% |
| 3 | Gemini 3.5 FlashGemini 3.5 Flash Model Card · May 2026· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 57.9% |
| 4 | Gemini 3.5 FlashOpus 4.8 System Card capability summary · adaptive thinking · max effort · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 57.9% |
| 5 | Gemini 3.5 Flash· LLM Stats data | 57.9% |
| 6 | Muse Spark 1.1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 57.2% |
| 7 | Muse Spark 1.1Meta Model API · xhigh · Vals AI result· Reported by Meta, Meta Superintelligence Labs· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 57.2% |
| 8 | Claude Fable 5· LLM Stats data | 56.3% |
| 9 | Mistral Large 4· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 54.7% |
| 10 | Kimi K3max effort · temp 1.0 · top_p 1.0· Reported by Moonshot AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 54.4% |
| 11 | Claude Opus 4.8· LLM Stats data | 53.9% |
| 12 | Claude Opus 4.8max · Vals AI result· Reported by Meta, Meta Superintelligence Labs· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 53.9% |
| 13 | Claude Opus 4.8Opus 4.8 System Card capability summary · adaptive thinking · max effort · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 53.9% |
| 14 | GPT 5.5Gemini 3.5 Flash Model Card · May 2026· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 51.8% |
| 15 | GPT 5.5Opus 4.8 System Card capability summary · adaptive thinking · max effort · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 51.8% |
| 16 | GPT 5.5xhigh · Vals AI result· Reported by Meta, Meta Superintelligence Labs· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 51.8% |
| 17 | GPT 5.5· LLM Stats data | 51.8% |
| 18 | Claude Opus 4.7· LLM Stats data | 51.5% |
| 19 | Claude Opus 4.7Gemini 3.5 Flash Model Card · May 2026· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 51.5% |
| 20 | Claude Opus 4.7Opus 4.8 System Card capability summary · adaptive thinking · max effort · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 51.5% |
| 21 | Claude Sonnet 4.6· LLM Stats data | 51% |
| 22 | Claude Sonnet 4.6Gemini 3.5 Flash Model Card · May 2026· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 51% |
| 23 | Qwen3.7 Max· LLM Stats data | 48.4% |
| 24 | MiniMax M3· LLM Stats data | 48.3% |
| 25 | GPT-5.4 mini· LLM Stats data | 45.4% |
| 26 | Kimi K2.6· LLM Stats data | 44.9% |
| 27 | GLM 5.1· LLM Stats data | 44.8% |
| 28 | Gemini 3.1 Pro PreviewGemini 3.5 Flash Model Card · May 2026· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 43% |
| 29 | Gemini 3.1 Pro Previewhigh · Vals AI result· Reported by Meta, Meta Superintelligence Labs· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 43% |
| 30 | Gemini 3.1 Pro PreviewOpus 4.8 System Card capability summary · adaptive thinking · max effort · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 43% |
| 31 | Gemini 3.1 Pro Preview· LLM Stats data | 43% |
| 32 | Gemini 3 FlashGemini 3.5 Flash Model Card · May 2026· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 42.6% |
| 33 | Gemini 3 Flash· LLM Stats data | 42.6% |
| 34 | MiMo V2.5 Pro· LLM Stats data | 41.5% |
| 35 | Qwen3.6 Plus· LLM Stats data | 40.9% |
| 36 | GPT 5.4 Nano· LLM Stats data | 38.2% |
| 37 | Qwen3.7 Plus· LLM Stats data | 38.2% |
| 38 | Grok 4.3· LLM Stats data | 37.7% |
| 39 | Nemotron 3 Ultra· LLM Stats data | 37.5% |
| 40 | MiMo V2.5· LLM Stats data | 36.7% |
| 41 | Mistral Medium 3.5· LLM Stats data | 32.1% |
| 42 | Claude Haiku 4.5· LLM Stats data | 31% |
| 43 | Gemini 3.1 Flash Lite· LLM Stats data | 30% |
| 44 | Grok 4.20reasoning, 0309, beta· LLM Stats data | 28.5% |
| 45 | MiniMax M2.7· LLM Stats data | 27.9% |
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.