Capabilities leaderboard across 11 benchmarks| Rank | Model | Harness | Provider | Overall score |
|---|
| 1 | GPT-6 Astra | Pi | OpenAI | 51.3% |
|---|
| 2 | GPT-6 Astra | Codex | OpenAI | 51.1% |
|---|
| 3 | Claude Opus 5 | Claude Code | Anthropic | 48.6% |
|---|
| 4 | Claude Opus 5 | Pi | Anthropic | 48.2% |
|---|
| 5 | Grok 4.6 | Grok Build | SpaceXAI | 48.2% |
|---|
| 6 | GPT-5.6 Sol | Pi | OpenAI | 47.5% |
|---|
| 7 | GPT-5.6 Sol | Codex | OpenAI | 45.9% |
|---|
| 8 | DeepSeek V4.1 Flash | Pi | DeepSeek | 45.0% |
|---|
| 9 | Grok 4.6 | Pi | SpaceXAI | 44.8% |
|---|
| 10 | Claude Opus 4.8 | Pi | Anthropic | 44.8% |
|---|
| 11 | Claude Opus 4.8 | Claude Code | Anthropic | 42.8% |
|---|
| 12 | GPT-5.6 Terra | Pi | OpenAI | 42.4% |
|---|
| 13 | Gemini 3.5 Flash | Pi | Google | 41.7% |
|---|
| 14 | GPT-5.6 Terra | Codex | OpenAI | 41.5% |
|---|
| 15 | Grok 4.5 | Pi | SpaceXAI | 41.3% |
|---|
| 16 | GPT-5.5 | Pi | OpenAI | 40.1% |
|---|
| 17 | Claude Opus 4.7 | Claude Code | Anthropic | 39.8% |
|---|
| 18 | Claude Sonnet 5 | Pi | Anthropic | 39.5% |
|---|
| 19 | Claude Sonnet 5 | Claude Code | Anthropic | 39.1% |
|---|
| 20 | GPT-5.5 | Codex | OpenAI | 38.7% |
|---|
| 21 | Kimi K3 | Pi | Moonshot | 38.1% |
|---|
| 22 | GPT-5.6 Luna | Pi | OpenAI | 33.8% |
|---|
| 23 | GPT-5.6 Luna | Codex | OpenAI | 32.2% |
|---|
Included benchmarks
- SpatialBench — Omics; published score: Verified subset (115 evaluations); Full benchmark: 159 evaluations
- SpatialBench-Long — Omics; published score: Verified subset (22 evaluations); Full benchmark: 24 evaluations
- scBench — Omics; published score: Full benchmark (195 evaluations)
- scBench-Long — Omics; published score: Verified merged subset (22 evaluations); Original full release: 21 evaluations
- EpiBench — Omics; published score: Full benchmark (106 evaluations)
- VariantBench — Omics; published score: Full benchmark (118 evaluations)
- TxBench-Antibody-Discovery — Therapeutics; published score: Full benchmark (100 evaluations)
- TxBench-Preclinical-Pharmacology — Therapeutics; published score: Full benchmark (100 evaluations)
- TxBench-Oligo-Discovery — Therapeutics; published score: Full benchmark (113 evaluations)
- BioSecBench-Surveillance — Biosecurity; published score: Full benchmark (102 evaluations)
- BioSecBench-Refusal — Biosecurity; published score: Full benchmark (107 evaluations)
- BioSecBench-Function — Biosecurity; published score: Full benchmark (111 evaluations)