# LFM2.5-24B-A2B Benchmark Scores & Performance

> LFM2.5-24B-A2B by LiquidAI scores 37.8/100 overall, ranking #145 out of 898 AI models.

Canonical page: https://benchlm.ai/models/lfm2-5-24b-a2b

Last updated: October 13, 2026

## Model Details

| Property | Value |
|----------|-------|
| Creator | LiquidAI |
| Source Type | Open Weight |
| Reasoning Type | Reasoning |
| Context Window | 131K |
| Official model card | [Liquid AI LFM2.5-24B-A2B documentation](https://docs.liquid.ai/lfm/models/lfm25-24b-a2b) |
| Overall Score | 37.8/100 |
| Overall Rank | #145 of 898 |

## Family & Coverage

- Family: LFM2.5-24B-A2B
- Variant: reasoning
- Benchmarks covered: 14 of 628
- Related earlier model: [LFM2-24B-A2B](/models/lfm2-24b-a2b)
- Coverage note: BenchLM currently has partial benchmark coverage for this model, so the overall score is conservative.

## Agentic Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [AutomationBench](/benchmarks/automationbench) | 27.7% | [Liquid AI: LFM2.5-24B-A2B launch post](https://www.liquid.ai/blog/lfm2-5-24b-a2b) |
| [BFCL v4](/benchmarks/bfcl-v4) | 65.0% | [Liquid AI: LFM2.5-24B-A2B launch post](https://www.liquid.ai/blog/lfm2-5-24b-a2b) |
| [τ³-bench results](/benchmarks/tau3-bench) | 8.0% | [Liquid AI: LFM2.5-24B-A2B launch post](https://www.liquid.ai/blog/lfm2-5-24b-a2b) |
| [PinchBench](/benchmarks/pinchbench) | 83.2% | [Liquid AI: LFM2.5-24B-A2B launch post](https://www.liquid.ai/blog/lfm2-5-24b-a2b) |
| [WebArena-Verified](/benchmarks/webarena-verified) | 26.8% | [Liquid AI: LFM2.5-24B-A2B launch post](https://www.liquid.ai/blog/lfm2-5-24b-a2b) |

## Coding Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [SciCode](/benchmarks/scicode) | 32.7% | [Liquid AI: LFM2.5-24B-A2B launch post](https://www.liquid.ai/blog/lfm2-5-24b-a2b) |
| [SWE-bench Verified](/benchmarks/swe-bench-verified) | 51% | [Liquid AI: LFM2.5-24B-A2B launch post](https://www.liquid.ai/blog/lfm2-5-24b-a2b) |

## Knowledge Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [GPQA](/benchmarks/gpqa) | 74.4% | [Liquid AI: LFM2.5-24B-A2B launch post](https://www.liquid.ai/blog/lfm2-5-24b-a2b) |
| [GPQA-D](/benchmarks/gpqa-diamond) | 74.4% | [Liquid AI: LFM2.5-24B-A2B launch post](https://www.liquid.ai/blog/lfm2-5-24b-a2b) |
| [HLE](/benchmarks/hle) | 7.4% | [Liquid AI: LFM2.5-24B-A2B launch post](https://www.liquid.ai/blog/lfm2-5-24b-a2b) |
| [AA-Omniscience Index](/benchmarks/aaomniscienceindex) | -24.9% | [Liquid AI: LFM2.5-24B-A2B launch post](https://www.liquid.ai/blog/lfm2-5-24b-a2b) |

## Instruction Following Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [IFEval](/benchmarks/ifeval) | 96.8% | [Liquid AI: LFM2.5-24B-A2B launch post](https://www.liquid.ai/blog/lfm2-5-24b-a2b) |
| [IFBench](/benchmarks/ifbench) | 74.7% | [Liquid AI: LFM2.5-24B-A2B launch post](https://www.liquid.ai/blog/lfm2-5-24b-a2b) |

## Mathematics Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [MATH-500](/benchmarks/math-500) | 96.0% | [Liquid AI: LFM2.5-24B-A2B launch post](https://www.liquid.ai/blog/lfm2-5-24b-a2b) |

## Other LiquidAI Models

- [LFM2.5-2.6B](/models/lfm2-5-2-6b) - Score: 31.09
- [LFM2.5-8B-A1B](/models/lfm2-5-8b-a1b) - Score: 30.02
- [LFM2-24B-A2B](/models/lfm2-24b-a2b) - Score: 20.59
- [LFM2.5-1.2B-Thinking](/models/lfm2-5-1-2b-thinking) - Score: 19.37
- [LFM2.5-1.2B-Instruct](/models/lfm2-5-1-2b-instruct) - Score: 19.12
- [LFM2.5-350M](/models/lfm2-5-350m) - Score: not computed
- [LFM2.5-VL-450M](/models/lfm2-5-vl-450m) - Score: not computed
- [LFM2.5-230M](/models/lfm2-5-230m) - Score: not computed
- [LFM2.5-VL-3B](/models/lfm2-5-vl-3b) - Score: not computed
- [LFM2.5-VL-1.6B-Extract](/models/lfm2-5-vl-1-6b-extract) - Score: not computed
- [LFM2.5-VL-450M-Extract](/models/lfm2-5-vl-450m-extract) - Score: not computed
- [d1-3B](/models/d1-3b) - Score: not computed
- [d1-omni-600M](/models/d1-omni-600m) - Score: not computed
- [LFM2.5-ColBERT-350M](/models/lfm2-5-colbert-350m) - Score: not computed
- [LFM2.5-Embedding-350M](/models/lfm2-5-embedding-350m) - Score: not computed
- [LFM2.5-1.2B-JP-202606](/models/lfm2-5-1-2b-jp-202606) - Score: not computed