# LFM2.5-230M vs MAI-Thinking-1

> Side-by-side benchmark comparison for LFM2.5-230M and MAI-Thinking-1 across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/lfm2-5-230m-vs-mai-thinking-1
- Last updated: September 27, 2026

- Shared sourced benchmarks: 4
- HTML indexing: indexable
- Ranking lane: BenchAlign v5.7

## Quick Verdict

Pick MAI-Thinking-1 if you want the stronger benchmark profile. LFM2.5-230M only makes more sense when its price, context window, or workload-specific category wins matter more than the overall score.

## Summary

- MAI-Thinking-1 leads overall 51 to 34.
- The clearest category separation is in instruction following, where the averages are 10.4 for LFM2.5-230M and 95.4 for MAI-Thinking-1.
- The biggest single benchmark swing is MMLU-Pro in Knowledge, with scores of 20.3% and 85%.
- MAI-Thinking-1 also has the larger context window at 256K.

## Model Snapshot

| Property | LFM2.5-230M | MAI-Thinking-1 |
|----------|----------|----------|
| Creator | LiquidAI | Microsoft |
| Type | Open Weight | Proprietary |
| Reasoning | Non-Reasoning | Reasoning |
| Context | 32K | 256K |
| Overall Score | null | null |
| Benchmarks Covered | 6 | 14 |
| Pricing (input/output) | $0.00 / $0.00 | N/A |

## Category Breakdown

### Agentic

- Winner: Not comparable
- LFM2.5-230M public-lane score: Coming soon
- MAI-Thinking-1 public-lane score: Coming soon

| Benchmark | LFM2.5-230M | MAI-Thinking-1 | Winner |
|-----------|-----------|-----------|--------|
| BFCL v4 | 21.0% | Coming soon | Coming soon |
| Terminal-Bench 2.0 | Coming soon | 46% | Coming soon |

### Coding

- Winner: Not comparable
- LFM2.5-230M public-lane score: Coming soon
- MAI-Thinking-1 public-lane score: Coming soon

| Benchmark | LFM2.5-230M | MAI-Thinking-1 | Winner |
|-----------|-----------|-----------|--------|
| LiveCodeBench v6 | Coming soon | 87.7% | Coming soon |
| SWE-bench Verified | Coming soon | 73.5% | Coming soon |
| SWE-bench Pro | Coming soon | 52.8% | Coming soon |
| Terminal-Bench 2.0 | Coming soon | 46.0% | Coming soon |

### Reasoning

- Winner: Not comparable
- LFM2.5-230M public-lane score: Coming soon
- MAI-Thinking-1 public-lane score: Coming soon

| Benchmark | LFM2.5-230M | MAI-Thinking-1 | Winner |
|-----------|-----------|-----------|--------|
| Graphwalks BFS 128K | Coming soon | 90% | Coming soon |

### Knowledge

- Winner: Not comparable
- LFM2.5-230M public-lane score: Coming soon
- MAI-Thinking-1 public-lane score: Coming soon

| Benchmark | LFM2.5-230M | MAI-Thinking-1 | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 25.4% | 84.2% | MAI-Thinking-1 |
| GPQA-D | 25.4% | 84.2% | MAI-Thinking-1 |
| MMLU-Pro | 20.3% | 85% | MAI-Thinking-1 |
| SimpleQA | Coming soon | 31% | Coming soon |

### Instruction Following

- Winner: MAI-Thinking-1
- LFM2.5-230M public-lane score: 10.4 (#123/124)
- MAI-Thinking-1 public-lane score: 95.4 (#1/124)

| Benchmark | LFM2.5-230M | MAI-Thinking-1 | Winner |
|-----------|-----------|-----------|--------|
| IFEval | 71.7% | Coming soon | Coming soon |
| IFBench | 38.4% | 85% | MAI-Thinking-1 |

### Mathematics

- Winner: Not comparable
- LFM2.5-230M public-lane score: Coming soon
- MAI-Thinking-1 public-lane score: 72.8 (Unranked · 3 rankable rows)

| Benchmark | LFM2.5-230M | MAI-Thinking-1 | Winner |
|-----------|-----------|-----------|--------|
| AIME 2025 | Coming soon | 97% | Coming soon |
| AIME26 | Coming soon | 94.5% | Coming soon |
| HMMT Feb 2026 | Coming soon | 84.9% | Coming soon |

## FAQ

### Which is better overall, LFM2.5-230M or MAI-Thinking-1?

MAI-Thinking-1 is ahead overall on BenchLM right now.

### Where is the biggest gap between LFM2.5-230M and MAI-Thinking-1?

The widest category gap is in instruction following, where the averages are 10.4 for LFM2.5-230M and 95.4 for MAI-Thinking-1.

### How many benchmarks does LFM2.5-230M cover on BenchLM?

LFM2.5-230M currently has 6 sourced benchmark scores on BenchLM.

### How many benchmarks does MAI-Thinking-1 cover on BenchLM?

MAI-Thinking-1 currently has 14 sourced benchmark scores on BenchLM.

## Related Comparisons

- [LFM2.5-230M vs GPT-6 Astra](/compare/gpt-6-astra-vs-lfm2-5-230m)
- [MAI-Thinking-1 vs GPT-6 Astra](/compare/gpt-6-astra-vs-mai-thinking-1)
- [LFM2.5-230M vs Claude Opus 5.5](/compare/claude-opus-5-5-vs-lfm2-5-230m)
- [MAI-Thinking-1 vs Claude Opus 5.5](/compare/claude-opus-5-5-vs-mai-thinking-1)

## Explore More

- [LFM2.5-230M profile](/models/lfm2-5-230m)
- [MAI-Thinking-1 profile](/models/mai-thinking-1)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)