# LFM2.5-230M vs LFM2.5-8B-A1B

> Side-by-side benchmark comparison for LFM2.5-230M and LFM2.5-8B-A1B across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/lfm2-5-230m-vs-lfm2-5-8b-a1b
- Last updated: September 27, 2026

- Shared sourced benchmarks: 3
- HTML indexing: indexable
- Ranking lane: BenchAlign v5.7

## Quick Verdict

Evidence is too thin for a benchmark verdict: LFM2.5-230M and LFM2.5-8B-A1B share only 3 sourced benchmark rows.

## Summary

- Only 3 shared sourced benchmark rows are available. That is not enough for a broad capability verdict, so this mirror does not name an overall winner.
- LFM2.5-8B-A1B has the larger published context window at 128K.

## Model Snapshot

| Property | LFM2.5-230M | LFM2.5-8B-A1B |
|----------|----------|----------|
| Creator | LiquidAI | LiquidAI |
| Type | Open Weight | Open Weight |
| Reasoning | Non-Reasoning | Reasoning |
| Context | 32K | 128K |
| Independent public score (not head-to-head) | null | 31.87 |
| Benchmarks Covered | 6 | 7 |
| Pricing (input/output) | $0.00 / $0.00 | $0.00 / $0.00 |

## Category Breakdown

### Agentic

- Winner: Insufficient shared evidence
- LFM2.5-230M public-lane score: Coming soon
- LFM2.5-8B-A1B public-lane score: Coming soon

| Benchmark | LFM2.5-230M | LFM2.5-8B-A1B | Winner |
|-----------|-----------|-----------|--------|
| BFCL v4 | 21.0% | 49.7% | LFM2.5-8B-A1B |
| τ²-bench results | Coming soon | 88.1% | Coming soon |

### Reasoning

- Winner: Insufficient shared evidence
- LFM2.5-230M public-lane score: Coming soon
- LFM2.5-8B-A1B public-lane score: 22.1 (Unranked · 2 rankable rows)

### Knowledge

- Winner: Insufficient shared evidence
- LFM2.5-230M public-lane score: Coming soon
- LFM2.5-8B-A1B public-lane score: 27.2 (Supported · #144/158)

| Benchmark | LFM2.5-230M | LFM2.5-8B-A1B | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 25.4% | Coming soon | Coming soon |
| GPQA-D | 25.4% | Coming soon | Coming soon |
| MMLU-Pro | 20.3% | Coming soon | Coming soon |

### Instruction Following

- Winner: Insufficient shared evidence
- LFM2.5-230M public-lane score: 10.4 (#123/124)
- LFM2.5-8B-A1B public-lane score: 35.4 (#108/124)

| Benchmark | LFM2.5-230M | LFM2.5-8B-A1B | Winner |
|-----------|-----------|-----------|--------|
| IFEval | 71.7% | 91.8% | LFM2.5-8B-A1B |
| IFBench | 38.4% | 56.5% | LFM2.5-8B-A1B |

### Mathematics

- Winner: Insufficient shared evidence
- LFM2.5-230M public-lane score: Coming soon
- LFM2.5-8B-A1B public-lane score: 17.3 (Unranked · 3 rankable rows)

| Benchmark | LFM2.5-230M | LFM2.5-8B-A1B | Winner |
|-----------|-----------|-----------|--------|
| MATH-500 | Coming soon | 88.8% | Coming soon |
| AIME 2025 | Coming soon | 42.5% | Coming soon |
| AIME26 | Coming soon | 50.0% | Coming soon |

## FAQ

### Is there enough evidence to pick between LFM2.5-230M and LFM2.5-8B-A1B?

Not yet. The models share only 3 sourced benchmark rows, so this mirror reports the available measurements and metadata without naming a broad winner.

### What can I compare safely on this page?

Published pricing, context windows, model type, and individual shared benchmark rows are directly comparable. Independent overall scores and non-overlapping benchmark coverage are labeled separately and should not be read as a controlled head-to-head result.

## Related Comparisons

- [LFM2.5-230M vs GPT-6 Astra](/compare/gpt-6-astra-vs-lfm2-5-230m)
- [LFM2.5-8B-A1B vs GPT-6 Astra](/compare/gpt-6-astra-vs-lfm2-5-8b-a1b)
- [LFM2.5-230M vs Claude Opus 5.5](/compare/claude-opus-5-5-vs-lfm2-5-230m)
- [LFM2.5-8B-A1B vs Claude Opus 5.5](/compare/claude-opus-5-5-vs-lfm2-5-8b-a1b)

## Explore More

- [LFM2.5-230M profile](/models/lfm2-5-230m)
- [LFM2.5-8B-A1B profile](/models/lfm2-5-8b-a1b)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)