# ZAYA1-8B Benchmark Scores & Performance

> ZAYA1-8B by Zyphra has 11 source-displayable benchmark rows but no public overall score. It remains unranked.

Canonical page: https://benchlm.ai/models/zaya1-8b

Last updated: September 18, 2026

## Model Details

| Property | Value |
|----------|-------|
| Creator | Zyphra |
| Source Type | Open Weight |
| Reasoning Type | Reasoning |
| Context Window | 131K |
| Overall Score | Coming soon |
| Overall Rank | Unranked |

## Family & Coverage

- Family: ZAYA1
- Variant: 8b
- Benchmarks covered: 11 of 446
- Sibling models: [ZAYA1-74B-Preview](/models/zaya1-74b-preview), [ZAYA1-VL-8B](/models/zaya1-vl-8b)
- Coverage note: BenchLM currently has partial benchmark coverage for this model, so the overall score is conservative.

## Agentic Benchmarks

| Benchmark | Score |
|-----------|-------|
| [BFCL v4](/benchmarks/bfcl-v4) | 39.2% |

## Coding Benchmarks

| Benchmark | Score |
|-----------|-------|
| [LiveCodeBench v6](/benchmarks/livecodebench-v6) | 65.8% |

## Knowledge Benchmarks

| Benchmark | Score |
|-----------|-------|
| [GPQA](/benchmarks/gpqa) | 71% |
| [GPQA-D](/benchmarks/gpqa-diamond) | 71.0% |
| [MMLU-Pro](/benchmarks/mmlu-pro) | 74.2% |

## Instruction Following Benchmarks

| Benchmark | Score |
|-----------|-------|
| [IFEval](/benchmarks/ifeval) | 85.6% |
| [IFBench](/benchmarks/ifbench) | 52.6% |

## Mathematics Benchmarks

| Benchmark | Score |
|-----------|-------|
| [AIME26](/benchmarks/aime2026) | 89.1% |
| [HMMT Feb 2026](/benchmarks/hmmtfeb2026) | 71.6% |
| [IMOAnswerBench](/benchmarks/imoanswerbench) | 59.3% |
| [Apex](/benchmarks/apex) | 32.2% |

## Other Zyphra Models

- [ZAYA1-74B-Preview](/models/zaya1-74b-preview) - Score: not computed
- [ZAYA1-VL-8B](/models/zaya1-vl-8b) - Score: not computed
- [Zonos 2](/models/zonos-2) - Score: not computed