FinHopper
Agents research financial facts through up to six dependent searches. Intermediate answer keys show where an incorrect company, date, or indicator derails the search.
- Made by
- South China University of Technology and Tencent
- Released
- October 2026
- Kind
- Finance
- Level
- University
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- All public
- Measures
- Capability
- Subject
- Economics and business
- Language
- English
- Interaction
- Agentic
- Environment
- APIs and tools
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Agents, Law and finance
- Strongest model tested
- GPT-5.2 #51 on the capability index
- Last model added
- December 2025
- Scores refreshed
- November 20th, 2026
Claude Sonnet 4.5 leads with 67.3% among models released by December 2025 Claude Sonnet 4.5 leads with 67.3% among models released by December 2025
Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.
The newest model with a score here came out in December 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 1 Claude Sonnet 4.5 thinking, use tools 67.3%
- 2
Gemini 3 Pro thinking, use tools 65%
- 3 GPT-5.2 thinking, use tools 61.3%
- 4
Seed1.8 thinking, use tools 61.1%
- 5
DeepSeek-V3.2 thinking, use tools 58.1%
- 6 Kimi K2 Thinking thinking, use tools 57.5%
- 7 HunYuan-2.0 thinking, use tools 57.3%
- 8 Kimi K2 Sep 2025, use tools 54.6%
- 9 HunYuan-2.0-instruct use tools 52.7%
- 10 Grok 4.1 thinking, use tools 52.7%
- 11 Qwen-max-20250923 thinking, use tools 52.5%
- 12 GLM-4.7 thinking, use tools 49%
Source: The FinHopper paper
- 1 Claude Sonnet 4.5 thinking, use tools 67.3%
- 2
Gemini 3 Pro thinking, use tools 65%
- 3 GPT-5.2 thinking, use tools 61.3%
- 4
Seed1.8 thinking, use tools 61.1%
- 5
DeepSeek-V3.2 thinking, use tools 58.1%
- 6 Kimi K2 Thinking thinking, use tools 57.5%
- 7 Kimi K2 Sep 2025, use tools 54.6%
- 8 Grok 4.1 thinking, use tools 52.7%
- 9 GLM-4.7 thinking, use tools 49%
Source: The FinHopper paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Sonnet 4.5 | 67.3% | thinking, use tools | Sep 2025 | Official leaderboard |
| 2 | | 65% | thinking, use tools | Nov 2025 | Official leaderboard |
| 3 | GPT-5.2 | 61.3% | thinking, use tools | Dec 2025 | Official leaderboard |
| 4 | | 61.1% | thinking, use tools | Dec 2025 | Official leaderboard |
| 5 | | 58.1% | thinking, use tools | Dec 2025 | Official leaderboard |
| 6 | Kimi K2 Thinking | 57.5% | thinking, use tools | Nov 2025 | Official leaderboard |
| 7 | HunYuan-2.0 | 57.3% | thinking, use tools | Official leaderboard | |
| 8 | Kimi K2 Sep 2025 | 54.6% | use tools | Sep 2025 | Official leaderboard |
| 9 | HunYuan-2.0-instruct | 52.7% | use tools | Official leaderboard | |
| 10 | Grok 4.1 | 52.7% | thinking, use tools | Nov 2025 | Official leaderboard |
| 11 | Qwen-max-20250923 | 52.5% | thinking, use tools | Official leaderboard | |
| 12 | GLM-4.7 | 49% | thinking, use tools | Dec 2025 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Sonnet 4.5 | 67.3% | thinking, use tools | Sep 2025 | Official leaderboard |
| 2 | | 65% | thinking, use tools | Nov 2025 | Official leaderboard |
| 3 | GPT-5.2 | 61.3% | thinking, use tools | Dec 2025 | Official leaderboard |
| 4 | | 61.1% | thinking, use tools | Dec 2025 | Official leaderboard |
| 5 | | 58.1% | thinking, use tools | Dec 2025 | Official leaderboard |
| 6 | Kimi K2 Thinking | 57.5% | thinking, use tools | Nov 2025 | Official leaderboard |
| 7 | Kimi K2 Sep 2025 | 54.6% | use tools | Sep 2025 | Official leaderboard |
| 8 | Grok 4.1 | 52.7% | thinking, use tools | Nov 2025 | Official leaderboard |
| 9 | GLM-4.7 | 49% | thinking, use tools | Dec 2025 | Official leaderboard |
Rankings by part
We read 7 parts from the benchmark’s paper, each ranking models on one part of the test, such as 1hop, 2hop and 3hop. Claude Sonnet 4.5 leads 4 of them, and 3 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
1hop
9 models we track have a score here.
- 1
DeepSeek-V3.2 ↑4 88.5%
- 2 GPT-5.2 ↑1 86.8%
- 3 Claude Sonnet 4.5 ↓2 81.3%
- 4
Gemini 3 Pro ↓2 81.2%
- 5
Seed1.8 ↓1 79%
- 6 GLM-4.7 ↑3 76.1%
- 7 Kimi K2 Sep 2025 68%
- 8 Kimi K2 Thinking ↓2 67.2%
- 9 Grok 4.1 ↓1 66.2%
2hop
9 models we track have a score here.
- 1 Claude Sonnet 4.5 72.5%
- 2
DeepSeek-V3.2 ↑3 72.3%
- 3
Gemini 3 Pro ↓1 70.3%
- 4 GPT-5.2 ↓1 69.6%
- 5
Seed1.8 ↓1 63.4%
- 6 Kimi K2 Thinking 58.5%
- 7 Grok 4.1 ↑1 57.7%
- 8 Kimi K2 Sep 2025 ↓1 50.9%
- 9 GLM-4.7 47.4%
3hop
9 models we track have a score here.
- 1 Claude Sonnet 4.5 70.2%
- 2
Gemini 3 Pro 69.5%
- 3
DeepSeek-V3.2 ↑2 65.7%
- 4 Kimi K2 Thinking ↑2 61.1%
- 5
Seed1.8 ↓1 60.2%
- 6 Kimi K2 Sep 2025 ↑1 59.7%
- 7 GPT-5.2 ↓4 56%
- 8 Grok 4.1 53%
- 9 GLM-4.7 50.7%
4hop
9 models we track have a score here.
- 1
Seed1.8 ↑3 63.9%
- 2 Claude Sonnet 4.5 ↓1 63.5%
- 3
Gemini 3 Pro ↓1 60.1%
- 4 GPT-5.2 ↓1 53.4%
- 5 Kimi K2 Thinking ↑1 53.2%
- 6 Grok 4.1 ↑2 50.3%
- 7 Kimi K2 Sep 2025 49.6%
- 8 GLM-4.7 ↑1 43.6%
- 9
DeepSeek-V3.2 ↓4 34.8%
5hop
9 models we track have a score here.
- 1 Claude Sonnet 4.5 54.1%
- 2
Gemini 3 Pro 52.2%
- 3 Kimi K2 Thinking ↑3 48.2%
- 4
Seed1.8 46.6%
- 5 GPT-5.2 ↓2 45.8%
- 6 Kimi K2 Sep 2025 ↑1 44%
- 7 Grok 4.1 ↑1 35.3%
- 8 GLM-4.7 ↑1 22.7%
- 9
DeepSeek-V3.2 ↓4 10.7%
6hop
9 models we track have a score here.
- 1 Kimi K2 Thinking ↑5 38%
- 2 Kimi K2 Sep 2025 ↑5 36.1%
- 3 Claude Sonnet 4.5 ↓2 31.1%
- 4
Seed1.8 28.8%
- 5
Gemini 3 Pro ↓3 26.5%
- 6 Grok 4.1 ↑2 25.8%
- 7 GPT-5.2 ↓4 19.2%
- 8 GLM-4.7 ↑1 14.2%
- 9
DeepSeek-V3.2 ↓4 9.30%
Mac Avg.
9 models we track have a score here.
- 1 Claude Sonnet 4.5 62.1%
- 2
Gemini 3 Pro 59.6%
- 3
Seed1.8 ↑1 56.6%
- 4 GPT-5.2 ↓1 54.6%
- 5 Kimi K2 Thinking ↑1 54.4%
- 6 Kimi K2 Sep 2025 ↑1 51.4%
- 7 Grok 4.1 ↑1 48.1%
- 8
DeepSeek-V3.2 ↓3 46.5%
- 9 GLM-4.7 42.5%
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: The FinHopper paper
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| Claude Sonnet 4.5 | thinking, use tools | 67.3% | |||
| thinking, use tools | 65% | ||||
| Claude Sonnet 4.5 | no-thinking, use tools | 63.7% | |||
| no-thinking, use tools | 62.7% | ||||
| GPT-5.2 | thinking, use tools | 61.3% | |||
| thinking, use tools | 61.1% | ||||
| use tools | 59.2% | ||||
| GPT-5.2 | use tools | 58.5% | |||
| thinking, use tools | 58.1% | ||||
| Kimi K2 Thinking | thinking, use tools | 57.5% | |||
| no-thinking, use tools | 57.1% | ||||
| Kimi K2 Sep 2025 | use tools | 54.6% | |||
| Grok 4.1 | thinking, use tools | 52.7% | |||
| GLM-4.7 | thinking, use tools | 49% | |||
| GLM-4.7 | no-thinking, use tools | 48.8% | |||
| Grok 4.1 | use tools | 42.7% | |||
| GLM-4.7 | thinking, without tools | 8.10% | |||
| Kimi K2 Sep 2025 | without tools | 7.00% | |||
| GLM-4.7 | no-thinking, without tools | 6.40% | |||
| Grok 4.1 | without tools | 6.20% | |||
| no-thinking, without tools | 4.20% | ||||
| GPT-5.2 | thinking, without tools | 4.20% | |||
| Kimi K2 Thinking | thinking, without tools | 3.40% | |||
| thinking, without tools | 3.20% | ||||
| without tools | 3.20% | ||||
| Grok 4.1 | thinking, without tools | 3.00% | |||
| no-thinking, without tools | 2.80% | ||||
| Claude Sonnet 4.5 | thinking, without tools | 2.50% | |||
| thinking, without tools | 2.00% | ||||
| thinking, without tools | 1.80% | ||||
| GPT-5.2 | without tools | 1.40% | |||
| Claude Sonnet 4.5 | no-thinking, without tools | 1.00% |
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: The FinHopper paper
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| Claude Sonnet 4.5 | thinking, use tools | 67.3% | |||
| thinking, use tools | 65% | ||||
| Claude Sonnet 4.5 | no-thinking, use tools | 63.7% | |||
| no-thinking, use tools | 62.7% | ||||
| GPT-5.2 | thinking, use tools | 61.3% | |||
| thinking, use tools | 61.1% | ||||
| use tools | 59.2% | ||||
| GPT-5.2 | use tools | 58.5% | |||
| thinking, use tools | 58.1% | ||||
| Kimi K2 Thinking | thinking, use tools | 57.5% | |||
| no-thinking, use tools | 57.1% | ||||
| Kimi K2 Sep 2025 | use tools | 54.6% | |||
| Grok 4.1 | thinking, use tools | 52.7% | |||
| GLM-4.7 | thinking, use tools | 49% | |||
| GLM-4.7 | no-thinking, use tools | 48.8% | |||
| Grok 4.1 | use tools | 42.7% | |||
| GLM-4.7 | thinking, without tools | 8.10% | |||
| Kimi K2 Sep 2025 | without tools | 7.00% | |||
| GLM-4.7 | no-thinking, without tools | 6.40% | |||
| Grok 4.1 | without tools | 6.20% | |||
| no-thinking, without tools | 4.20% | ||||
| GPT-5.2 | thinking, without tools | 4.20% | |||
| Kimi K2 Thinking | thinking, without tools | 3.40% | |||
| thinking, without tools | 3.20% | ||||
| without tools | 3.20% | ||||
| Grok 4.1 | thinking, without tools | 3.00% | |||
| no-thinking, without tools | 2.80% | ||||
| Claude Sonnet 4.5 | thinking, without tools | 2.50% | |||
| thinking, without tools | 2.00% | ||||
| thinking, without tools | 1.80% | ||||
| GPT-5.2 | without tools | 1.40% | |||
| Claude Sonnet 4.5 | no-thinking, without tools | 1.00% |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The FinHopper paper
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official FinHopper leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Law, finance and medicine
- MedXpertQA-MM In use Models answer expert-level medical exam questions that combine images with clinical records and test results. Scoring is accuracy.
Qwen 3.8 Max 80.4%
- HealthBench Professional In use 525 physician-authored tasks from real clinician chats (care consult, documentation, research) with physician-adjudicated rubrics. Claude Sonnet 5.5 69.2%
- HealthBench In use 5,000 realistic health conversations graded against physician-written rubrics. A length-adjusted variant penalizes verbosity. GPT-5 67.2%
- FinSearchComp In use Agents search the web to fetch live financial data, look up history, and run multi-step investigations, across 635 expert-written questions. Scored by answer accuracy.
Ling-3.1-flash 81.7%
- KorMedMCQA In use Models answer 7,469 multiple-choice questions from Korean licensing exams for doctors, nurses, pharmacists and dentists, held from 2012 to 2024. o1-preview 92.7%
- Harvey LAB-AA v1.1 In use Artificial Analysis's run of Harvey's Legal Agent Benchmark on Harvey's newer private set of 120 tasks across 24 practice areas. A deliverable with a material hallucination scores zero, and the headline is the share of judges who find every rubric criterion passed. GPT-6.2 Astra 15%
Data audit
9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 11 months ago, in December 2025, for GLM-4.7. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
- Pass Independent runs Scores from the benchmark’s paper (12 entries).
- Pass Models scored 9 models we track, plus 3 other entries on its leaderboard.
- Warning Newest score The newest came 11 months ago, in December 2025, for GLM-4.7.
- Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.2 (#51).
- Pass Headroom The best is 67.3%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on arxiv.org: 32 rows for models we track and 3 other entries at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.