# Agents-A1-4B Benchmark Scores & Performance

> Agents-A1-4B by InternScience has 11 source-displayable benchmark rows but no public overall score. It remains unranked.

Canonical page: https://benchlm.ai/models/agents-a1-4b

Last updated: September 10, 2026

## Model Details

| Property | Value |
|----------|-------|
| Creator | InternScience |
| Source Type | Open Weight |
| Reasoning Type | Reasoning |
| Context Window | 262K |
| Official model card | [InternScience Agents-A1-4B model card](https://huggingface.co/InternScience/Agents-A1-4B) |
| Overall Score | Coming soon |
| Overall Rank | Unranked |

## Family & Coverage

- Family: Agents-A1
- Variant: 4b
- Benchmarks covered: 11 of 434
- Sibling models: [Agents-A1](/models/agents-a1), [Agents-A1-F16-GGUF](/models/agents-a1-f16-gguf), [Agents-A1-FP8](/models/agents-a1-fp8), [Agents-A1-Q4_K_M-GGUF](/models/agents-a1-q4-k-m-gguf), [Agents-A1-Q8_0-GGUF](/models/agents-a1-q8-0-gguf)
- Coverage note: BenchLM currently has partial benchmark coverage for this model, so the overall score is conservative.

## Agentic Benchmarks

| Benchmark | Score |
|-----------|-------|
| [BrowseComp](/benchmarks/browsecomp) | 66.8% |
| [GAIA](/benchmarks/gaia) | 95.1% |
| [MLE-Bench Lite](/benchmarks/mle-bench-lite) | 22.7% |
| [τ²-bench results](/benchmarks/tau2-bench) | 78.2% |
| [VITA-Bench](/benchmarks/vitabench) | 40.3% |

## Coding Benchmarks

| Benchmark | Score |
|-----------|-------|
| [SciCode](/benchmarks/scicode) | 29.6% |
| [LiveCodeBench v6](/benchmarks/livecodebench-v6) | 59.6% |

## Reasoning Benchmarks

| Benchmark | Score |
|-----------|-------|
| [LongBench v2](/benchmarks/longbench-v2) | 52.1% |

## Knowledge Benchmarks

| Benchmark | Score |
|-----------|-------|
| [FrontierScience Research](/benchmarks/frontierscienceresearch) | 33.3% |

## Instruction Following Benchmarks

| Benchmark | Score |
|-----------|-------|
| [IFBench](/benchmarks/ifbench) | 69.1% |
| [IFEval](/benchmarks/ifeval) | 94.8% |

## Other InternScience Models

- [Agents-A1](/models/agents-a1) - Score: not computed
- [Agents-A1-F16-GGUF](/models/agents-a1-f16-gguf) - Score: not computed
- [Agents-A1-FP8](/models/agents-a1-fp8) - Score: not computed
- [Agents-A1-Q4_K_M-GGUF](/models/agents-a1-q4-k-m-gguf) - Score: not computed
- [Agents-A1-Q8_0-GGUF](/models/agents-a1-q8-0-gguf) - Score: not computed