# LFM2.5-2.6B vs LFM2.5-24B-A2B

> Side-by-side benchmark comparison for LFM2.5-2.6B and LFM2.5-24B-A2B across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/lfm2-5-2-6b-vs-lfm2-5-24b-a2b
- Last updated: October 13, 2026

- Shared sourced benchmarks: 4
- HTML indexing: indexable
- Ranking lane: BenchAlign v5.8

## Quick Verdict

LFM2.5-24B-A2B leads by point estimate. Conditional score ranges do not establish rank confidence. Choose using the category evidence and your own workload tests.

## Summary

- LFM2.5-24B-A2B has the higher overall point estimate 37.8 to 31.09. Conditional score ranges do not establish rank confidence.
- The clearest category separation is in instruction following, where the averages are 39.1 for LFM2.5-2.6B and 75.6 for LFM2.5-24B-A2B.
- The biggest single benchmark swing is IFBench in Instruction Following, with scores of 59.2% and 74.7%.
- LFM2.5-2.6B is the cheaper option on output tokens, which matters if you expect large responses or heavy interactive use.
- LFM2.5-24B-A2B also has the larger context window at 131K.

## Model Snapshot

| Property | LFM2.5-2.6B | LFM2.5-24B-A2B |
|----------|----------|----------|
| Creator | LiquidAI | LiquidAI |
| Type | Open Weight | Open Weight |
| Reasoning | Reasoning | Reasoning |
| Context | 128K | 131K |
| Overall Score | 31.09 | 37.8 |
| Benchmarks Covered | 7 | 13 |
| Pricing (input/output) | $0.00 / $0.00 | $0.10 / $1.00 |

## Category Breakdown

### Agentic

- Winner: Directional only
- LFM2.5-2.6B public-lane score: 16.3 (Estimated · #109/124)
- LFM2.5-24B-A2B public-lane score: 30.2 (Estimated · #88/124)

| Benchmark | LFM2.5-2.6B | LFM2.5-24B-A2B | Winner |
|-----------|-----------|-----------|--------|
| BFCL v4 | 56.9% | 65.0% | LFM2.5-24B-A2B |
| τ³-bench results | 5.7% | 8.0% | LFM2.5-24B-A2B |
| Claw-Eval | 62.9% | Coming soon | Coming soon |
| PinchBench | 68.2% | 83.2% | LFM2.5-24B-A2B |
| AutomationBench | Coming soon | 27.7% | Coming soon |
| WebArena-Verified | Coming soon | 26.8% | Coming soon |

### Coding

- Winner: Directional only
- LFM2.5-2.6B public-lane score: 12.7 (Estimated · #138/149)
- LFM2.5-24B-A2B public-lane score: 28.5 (Estimated · #98/149)

| Benchmark | LFM2.5-2.6B | LFM2.5-24B-A2B | Winner |
|-----------|-----------|-----------|--------|
| LiveCodeBench v6 | 59.4% | Coming soon | Coming soon |
| SciCode | Coming soon | 32.7% | Coming soon |
| SWE-bench Verified | Coming soon | 51% | Coming soon |

### Reasoning

- Winner: Not comparable
- LFM2.5-2.6B public-lane score: 29.8 (Unranked · 2 rankable rows)
- LFM2.5-24B-A2B public-lane score: Coming soon

### Knowledge

- Winner: Directional only
- LFM2.5-2.6B public-lane score: 26.7 (Estimated · #165/181)
- LFM2.5-24B-A2B public-lane score: 40.8 (Estimated · #112/181)

| Benchmark | LFM2.5-2.6B | LFM2.5-24B-A2B | Winner |
|-----------|-----------|-----------|--------|
| GPQA | Coming soon | 74.4% | Coming soon |
| GPQA-D | Coming soon | 74.4% | Coming soon |
| HLE | Coming soon | 7.4% | Coming soon |

### Instruction Following

- Winner: LFM2.5-24B-A2B
- LFM2.5-2.6B public-lane score: 39.1 (#106/128)
- LFM2.5-24B-A2B public-lane score: 75.6 (#61/128)

| Benchmark | LFM2.5-2.6B | LFM2.5-24B-A2B | Winner |
|-----------|-----------|-----------|--------|
| IFBench | 59.2% | 74.7% | LFM2.5-24B-A2B |
| IFEval | Coming soon | 96.8% | Coming soon |

### Mathematics

- Winner: Not comparable
- LFM2.5-2.6B public-lane score: Coming soon
- LFM2.5-24B-A2B public-lane score: Coming soon

| Benchmark | LFM2.5-2.6B | LFM2.5-24B-A2B | Winner |
|-----------|-----------|-----------|--------|
| AIME 2025 | 51.9% | Coming soon | Coming soon |
| MATH-500 | Coming soon | 96.0% | Coming soon |

## FAQ

### Which is better overall, LFM2.5-2.6B or LFM2.5-24B-A2B?

LFM2.5-24B-A2B has the higher overall point estimate. Conditional score ranges do not establish rank confidence.

### Where is the biggest gap between LFM2.5-2.6B and LFM2.5-24B-A2B?

The widest category gap is in instruction following, where the averages are 39.1 for LFM2.5-2.6B and 75.6 for LFM2.5-24B-A2B.

### How many benchmarks does LFM2.5-2.6B cover on BenchLM?

LFM2.5-2.6B currently has 7 sourced benchmark scores on BenchLM.

### How many benchmarks does LFM2.5-24B-A2B cover on BenchLM?

LFM2.5-24B-A2B currently has 13 sourced benchmark scores on BenchLM.

## Related Comparisons

- [LFM2.5-2.6B vs Claude Opus 5.5](/compare/claude-opus-5-5-vs-lfm2-5-2-6b)
- [LFM2.5-24B-A2B vs Claude Opus 5.5](/compare/claude-opus-5-5-vs-lfm2-5-24b-a2b)
- [LFM2.5-2.6B vs GPT-6.1 Sol](/compare/gpt-6-1-sol-vs-lfm2-5-2-6b)
- [LFM2.5-24B-A2B vs GPT-6.1 Sol](/compare/gpt-6-1-sol-vs-lfm2-5-24b-a2b)

## Explore More

- [LFM2.5-2.6B profile](/models/lfm2-5-2-6b)
- [LFM2.5-24B-A2B profile](/models/lfm2-5-24b-a2b)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)