# SciUniverse Level 1

> SciUniverse Level 1 tests whether models carry out scientific work across chemistry, biology, and materials science. Its 92 tasks include physical experiments, simulated lab work, and measurement analysis, with Pass@1 averaged equally across 17 task families.

Canonical page: https://benchlm.ai/benchmarks/sciuniverse

- Category: [Agentic](/agentic)
- Last updated: October 15, 2026

Evaluation results published by [C5R](https://c5r.net/sciuniverse/). Published Level 1 results and evaluation settings. Task contents and model responses are not included.

The October 15, 2026 snapshot keeps C5R's Level 1 results. The 17 task families receive equal overall weight; chemistry has 7 families, biology 5, materials 4, and cross-domain 1.

The rows evaluate models inside the C5R facility harness. We keep them display only, outside overall and category rankings. Model links are for navigation and do not add these scores to model profiles.

- [C5R results and task space](https://c5r.net/sciuniverse/)
- [C5R facility harness](https://c5r.net/#harness-title)

## About SciUniverse Level 1

- Year: 2026
- Tasks: 92 tasks in 17 task families
- Format: Scientific work through a facility harness; Pass@1
- Difficulty: Level 1 scientific tasks

Models direct equipment and human operators through the C5R facility harness. Level 1 covers work that scientists can complete in a few hours. We show the published overall and four domain scores, effort settings, and inference costs as a standalone table.

SciUniverse Level 1 is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (6 model-harness configurations)

| Source model | Evaluated identity | Effort | Overall Pass@1 | Chemistry (7 families) | Biology (5 families) | Materials (4 families) | Cross-domain (1 family) | Inference cost / attempt | Cost estimates |
|---|---|---|---|---|---|---|---|---|---|
| [Claude Fable 5.1](/models/claude-fable-5-1) | claude-fable-5-1 | xhigh | 45.3% | 37.4% | 40.0% | 68.8% | 34.4% | $40.61 | None |
| [GPT–6 Astra](/models/gpt-6-astra) | gpt-6-astra | xhigh | 32.5% | 23.3% | 60.0% | 20.6% | 6.3% | $52.37 | None |
| [Claude Opus 5](/models/claude-opus-5) | claude-opus-5 | xhigh | 30.5% | 12.1% | 31.1% | 68.1% | 6.3% | $46.31 | None |
| [Grok 4.6](/models/grok-4-6) | grok-4.6 | xhigh | 26.2% | 17.5% | 32.0% | 40.9% | 0.0% | $13.41 | None |
| [Gemini 3.8 Flash](/models/gemini-3-8-flash) | gemini-3.8-flash | high | 14.6% | 6.5% | 28.9% | 14.7% | 0.0% | $4.55 | 12 of 361 attempts |
| [GPT–5.6 Sol](/models/gpt-5-6-sol) | gpt-5.6-sol | xhigh | 9.4% | 3.1% | 15.6% | 15.3% | 0.0% | $16.53 | None |

Overall Pass@1 gives each of the 17 task families equal weight. Domain columns average their own families. All rows use the C5R facility harness. Model links are for navigation; these results stay outside overall and category rankings.

Inference cost is per task attempt, including failures and source estimates, with the same family weights as scores. Lab and labor costs are excluded. C5R captured the source tariffs on September 9, 2026. 

## FAQ

### What does SciUniverse Level 1 measure?

SciUniverse Level 1 tests whether models carry out scientific work across chemistry, biology, and materials science. Its 92 tasks include physical experiments, simulated lab work, and measurement analysis, with Pass@1 averaged equally across 17 task families.

### Which model leads the published SciUniverse Level 1 snapshot?

Claude Fable 5.1 currently leads the published SciUniverse Level 1 snapshot with a score of 45.3%.

### How many models are evaluated on SciUniverse Level 1?

The October 15, 2026 contains 6 model-harness configurations on 92 tasks in 17 task families.

### Does SciUniverse Level 1 affect BenchLM's overall score?

Not directly. SciUniverse Level 1 is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.