# Claude Sonnet 4.6 vs LFM2.5-2.6B

> Side-by-side benchmark comparison for Claude Sonnet 4.6 and LFM2.5-2.6B across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/claude-sonnet-4-6-vs-lfm2-5-2-6b
- Last updated: September 18, 2026

- Shared sourced benchmarks: 1
- HTML indexing: indexable
- Ranking lane: BenchAlign v5

## Quick Verdict

Evidence is too thin for a benchmark verdict: Claude Sonnet 4.6 and LFM2.5-2.6B share only 1 sourced benchmark row.

## Summary

- Only 1 shared sourced benchmark row is available. That is not enough for a broad capability verdict, so this mirror does not name an overall winner.
- LFM2.5-2.6B has the lower published output-token price.
- Claude Sonnet 4.6 has the larger published context window at 200K.

## Model Snapshot

| Property | Claude Sonnet 4.6 | LFM2.5-2.6B |
|----------|----------|----------|
| Creator | Anthropic | LiquidAI |
| Type | Proprietary | Open Weight |
| Reasoning | Non-Reasoning | Reasoning |
| Context | 200K | 128K |
| Independent public score (not head-to-head) | 62.86 | 39.9 |
| Benchmarks Covered | 26 | 7 |
| Pricing (input/output) | $3.00 / $15.00 | $0.00 / $0.00 |

## Category Breakdown

### Agentic

- Winner: Insufficient shared evidence
- Claude Sonnet 4.6 public-lane score: 43.8 (Supported · #95/154)
- LFM2.5-2.6B public-lane score: 37.2 (Estimated · #131/154)

| Benchmark | Claude Sonnet 4.6 | LFM2.5-2.6B | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 2.0 | 59.1% | Coming soon | Coming soon |
| OSWorld-Verified | 72.1% | Coming soon | Coming soon |
| Claw-Eval | 67.8% | 62.9% | Claude Sonnet 4.6 |
| CyberGym | 65.2% | Coming soon | Coming soon |
| Gert Labs | 62.92% | Coming soon | Coming soon |
| OSWorld 2.0 | 8.3% | Coming soon | Coming soon |
| JobBench | 36.9% | Coming soon | Coming soon |
| Terminal-Bench 2.1 (Vals) | 57.3% | Coming soon | Coming soon |
| ApprenticeBench | 2% | Coming soon | Coming soon |
| BFCL v4 | Coming soon | 56.9% | Coming soon |
| τ³-bench results | Coming soon | 5.7% | Coming soon |
| PinchBench | Coming soon | 68.2% | Coming soon |

### Coding

- Winner: Insufficient shared evidence
- Claude Sonnet 4.6 public-lane score: 52.2 (Supported · #49/154)
- LFM2.5-2.6B public-lane score: 39.2 (Estimated · #125/154)

| Benchmark | Claude Sonnet 4.6 | LFM2.5-2.6B | Winner |
|-----------|-----------|-----------|--------|
| SWE-bench Verified | 79.6% | Coming soon | Coming soon |
| SWE-Rebench | 60.7% | Coming soon | Coming soon |
| React Native Evals | 80.6% | Coming soon | Coming soon |
| Vibe Code Bench | 51.48% | Coming soon | Coming soon |
| cursorBench31 | 48.8% | Coming soon | Coming soon |
| FrontierCode 1.1 Main | 24.3% | Coming soon | Coming soon |
| LiveCodeBench (Vals) | 82.1% | Coming soon | Coming soon |
| SWE-bench (Vals) | 77.4% | Coming soon | Coming soon |
| LiveCodeBench v6 | Coming soon | 59.4% | Coming soon |

### Multimodal & Grounded

- Winner: Insufficient shared evidence
- Claude Sonnet 4.6 public-lane score: 54.1 (#33/48)
- LFM2.5-2.6B public-lane score: Coming soon

| Benchmark | Claude Sonnet 4.6 | LFM2.5-2.6B | Winner |
|-----------|-----------|-----------|--------|
| CharXiv | 77.4% | Coming soon | Coming soon |

### Reasoning

- Winner: Insufficient shared evidence
- Claude Sonnet 4.6 public-lane score: 67.9 (Unranked · 2 rankable rows)
- LFM2.5-2.6B public-lane score: 24.9 (Unranked · 2 rankable rows)

### Knowledge

- Winner: Insufficient shared evidence
- Claude Sonnet 4.6 public-lane score: 55.7 (Supported · #51/184)
- LFM2.5-2.6B public-lane score: 37 (Estimated · #151/184)

| Benchmark | Claude Sonnet 4.6 | LFM2.5-2.6B | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 89.9% | Coming soon | Coming soon |
| SuperGPQA | 95% | Coming soon | Coming soon |
| MMLU-Pro | 79.2% | Coming soon | Coming soon |
| HLE | 49% | Coming soon | Coming soon |
| GPQA Diamond (Vals) | 85.6% | Coming soon | Coming soon |
| MMLU-Pro (Vals) | 87.3% | Coming soon | Coming soon |

### Instruction Following

- Winner: Insufficient shared evidence
- Claude Sonnet 4.6 public-lane score: 46.6 (#88/124)
- LFM2.5-2.6B public-lane score: 38.1 (#104/124)

| Benchmark | Claude Sonnet 4.6 | LFM2.5-2.6B | Winner |
|-----------|-----------|-----------|--------|
| IFBench | Coming soon | 59.2% | Coming soon |

### Mathematics

- Winner: Insufficient shared evidence
- Claude Sonnet 4.6 public-lane score: 48.8 (Unranked · 2 rankable rows)
- LFM2.5-2.6B public-lane score: Coming soon

| Benchmark | Claude Sonnet 4.6 | LFM2.5-2.6B | Winner |
|-----------|-----------|-----------|--------|
| FrontierMath v2 (Tiers 1-3) | 32.400% | Coming soon | Coming soon |
| FrontierMath v2 (Tier 4) | 8.300% | Coming soon | Coming soon |
| AIME 2025 | Coming soon | 51.9% | Coming soon |

## FAQ

### Is there enough evidence to pick between Claude Sonnet 4.6 and LFM2.5-2.6B?

Not yet. The models share only 1 sourced benchmark row, 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

- [Claude Sonnet 4.6 vs Claude Fable 5.1](/compare/claude-fable-5-1-vs-claude-sonnet-4-6)
- [LFM2.5-2.6B vs Claude Fable 5.1](/compare/claude-fable-5-1-vs-lfm2-5-2-6b)
- [Claude Sonnet 4.6 vs GPT-6 Astra](/compare/claude-sonnet-4-6-vs-gpt-6-astra)
- [LFM2.5-2.6B vs GPT-6 Astra](/compare/gpt-6-astra-vs-lfm2-5-2-6b)

## Explore More

- [Claude Sonnet 4.6 profile](/models/claude-sonnet-4-6)
- [LFM2.5-2.6B profile](/models/lfm2-5-2-6b)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)