# DeepSeek V3 vs LFM2.5-230M

> Side-by-side benchmark comparison for DeepSeek V3 and LFM2.5-230M across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/deepseek-v3-vs-lfm2-5-230m
- 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: DeepSeek V3 and LFM2.5-230M 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.
- LFM2.5-230M has the lower published output-token price.
- DeepSeek V3 has the larger published context window at 128K.

## Model Snapshot

| Property | DeepSeek V3 | LFM2.5-230M |
|----------|----------|----------|
| Creator | DeepSeek | LiquidAI |
| Type | Open Weight | Open Weight |
| Reasoning | Non-Reasoning | Non-Reasoning |
| Context | 128K | 32K |
| Independent public score (not head-to-head) | 31.9 | null |
| Benchmarks Covered | 6 | 6 |
| Pricing (input/output) | $0.27 / $1.10 | $0.00 / $0.00 |

## Category Breakdown

### Agentic

- Winner: Insufficient shared evidence
- DeepSeek V3 public-lane score: 11.1 (Estimated · #102/105)
- LFM2.5-230M public-lane score: Coming soon

| Benchmark | DeepSeek V3 | LFM2.5-230M | Winner |
|-----------|-----------|-----------|--------|
| BFCL v4 | Coming soon | 21.0% | Coming soon |

### Coding

- Winner: Insufficient shared evidence
- DeepSeek V3 public-lane score: 18.9 (Estimated · #123/135)
- LFM2.5-230M public-lane score: Coming soon

| Benchmark | DeepSeek V3 | LFM2.5-230M | Winner |
|-----------|-----------|-----------|--------|
| LiveCodeBench | 37.6% | Coming soon | Coming soon |
| SWE-bench Verified | 42% | Coming soon | Coming soon |

### Reasoning

- Winner: Insufficient shared evidence
- DeepSeek V3 public-lane score: 42.2 (Unranked · 2 rankable rows)
- LFM2.5-230M public-lane score: Coming soon

### Knowledge

- Winner: Insufficient shared evidence
- DeepSeek V3 public-lane score: 30 (Estimated · #130/158)
- LFM2.5-230M public-lane score: Coming soon

| Benchmark | DeepSeek V3 | LFM2.5-230M | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 59.1% | 25.4% | DeepSeek V3 |
| MMLU-Pro | 75.9% | 20.3% | DeepSeek V3 |
| GPQA-D | Coming soon | 25.4% | Coming soon |

### Instruction Following

- Winner: Insufficient shared evidence
- DeepSeek V3 public-lane score: 38.2 (#103/124)
- LFM2.5-230M public-lane score: 10.4 (#123/124)

| Benchmark | DeepSeek V3 | LFM2.5-230M | Winner |
|-----------|-----------|-----------|--------|
| IFEval | 86.1% | 71.7% | DeepSeek V3 |
| IFBench | Coming soon | 38.4% | Coming soon |

### Mathematics

- Winner: Insufficient shared evidence
- DeepSeek V3 public-lane score: 26 (Unranked · 1 rankable row)
- LFM2.5-230M public-lane score: Coming soon

| Benchmark | DeepSeek V3 | LFM2.5-230M | Winner |
|-----------|-----------|-----------|--------|
| FrontierMath v2 (Tiers 1-3) | 1.724% | Coming soon | Coming soon |

## FAQ

### Is there enough evidence to pick between DeepSeek V3 and LFM2.5-230M?

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

- [DeepSeek V3 vs GPT-6 Astra](/compare/deepseek-v3-vs-gpt-6-astra)
- [LFM2.5-230M vs GPT-6 Astra](/compare/gpt-6-astra-vs-lfm2-5-230m)
- [DeepSeek V3 vs Claude Opus 5.5](/compare/claude-opus-5-5-vs-deepseek-v3)
- [LFM2.5-230M vs Claude Opus 5.5](/compare/claude-opus-5-5-vs-lfm2-5-230m)

## Explore More

- [DeepSeek V3 profile](/models/deepseek-v3)
- [LFM2.5-230M profile](/models/lfm2-5-230m)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)