# Mellum2-12B-A2.5B-Thinking Benchmark Scores & Performance

> Mellum2-12B-A2.5B-Thinking by JetBrains has 6 source-displayable benchmark rows but no public overall score. It remains unranked.

Canonical page: https://benchlm.ai/models/mellum2-12b-a2-5b-thinking

Last updated: September 4, 2026

## Model Details

| Property | Value |
|----------|-------|
| Creator | JetBrains |
| Source Type | Open Weight |
| Reasoning Type | Reasoning |
| Context Window | 128K |
| Overall Score | Coming soon |
| Overall Rank | Unranked |

## Family & Coverage

- Family: Mellum2 12B-A2.5B
- Variant: thinking
- Benchmarks covered: 6 of 422
- Sibling models: [Mellum2-12B-A2.5B-Instruct](/models/mellum2-12b-a2-5b-instruct)
- 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) | 45.6% |

## Coding Benchmarks

| Benchmark | Score |
|-----------|-------|
| [LiveCodeBench v6](/benchmarks/livecodebench-v6) | 69.9% |

## Knowledge Benchmarks

| Benchmark | Score |
|-----------|-------|
| [MMLU-Redux](/benchmarks/mmluredux) | 86.2% |
| [GPQA](/benchmarks/gpqa) | 57.6% |
| [GPQA-D](/benchmarks/gpqa-diamond) | 57.6% |

## Instruction Following Benchmarks

| Benchmark | Score |
|-----------|-------|
| [IFEval](/benchmarks/ifeval) | 76.5% |

## Other JetBrains Models

- [Mellum2-12B-A2.5B-Instruct](/models/mellum2-12b-a2-5b-instruct) - Score: not computed