Benchmarks
Benchmark results
TrialMind against general-purpose models and published methods, with the peer-reviewed source behind every result.
- Benchmarks
- 31
- Benchmarks
- Task families
- 8
- Task families
- Published sources
- 8
- Published sources
Results by task
Benchmarks grouped by task type, from biomedical reasoning to digital twins. Each panel links the paper the results come from.
Knowledge-intensive biomedical reasoning
Source: DeepEvidence, arXiv 2025 ›HLE-Medicine
12×- TrialMind40
- Biomni20
- ToolUniverse6.7
- Sonnet-4.510
- GPT-53.3
LabBench-LitQA2
1.7×- TrialMind80
- GPT-548
- Sonnet-4.540
- Biomni32
- ToolUniverse24
SuperGPQA-Hard
1.6×- TrialMind47.1
- Sonnet-4.543.6
- Biomni40.7
- ToolUniverse37.2
- GPT-529.4
Target Identification
1.5×- TrialMind68
- Biomni56
- GPT-544
- ToolUniverse40
- Sonnet-4.538
Mechanism of Action (MOA) & Pathway Reasoning
1.6×- TrialMind72
- LLMs52
- Biomni44
- ToolUniverse44
In Vivo Flux Response
1.5×- TrialMind80
- Biomni68
- LLMs60
- ToolUniverse52
Scores and improvement figures as reported in each source.
Compare TrialMind on your own task
Bring a question you have already answered. Compare the result to your own.