Compare / LFM2.5 24B-A2B vs Mistral Small 3.2
LFM2.5 24B-A2B vs Mistral Small 3.2
Updated November 2026
Verdict
LFM2.5 24B-A2B scores higher on the Small LM Index: 61.4 vs 38.6. On the 1 benchmark both labs report, LFM2.5 24B-A2B wins 1 and Mistral Small 3.2 wins 0. Both need about 16.9 GB of memory at 4-bit quantization. Mistral Small 3.2 is cheaper through an API: $0.11 per million tokens against $0.21.
Pick LFM2.5 24B-A2B if…
- you want the strongest model overall
Pick Mistral Small 3.2 if…
- API cost matters most
- you need a permissive license (Apache-2.0)
- you need image input
- you want the more widely used model (more quantized builds and community support)
Specifications
| LFM2.5 24B-A2B | Mistral Small 3.2 | |
|---|---|---|
| Small LM Index | 61.4 (likely 57–65) | 38.6 (likely 29–49) |
| Lab | Liquid AI · United States | Mistral AI · Europe |
| Released | 2026-10-13 | 2025-06-20 |
| Parameters | 24B (2B active, MoE) | 24B (dense) |
| Memory at 4-bit | ~15.7 GB | ~16.9 GB |
| Context window | 128K tokens | 128K tokens |
| Input | text | text, image |
| Reasoning | No thinking | No thinking |
| License | LFM Open License | Apache-2.0 |
| API price / 1M tokens | $0.21 ($0.050 in, $0.70 out) | $0.11 ($0.075 in, $0.20 out) |
| Hugging Face downloads (30 days) | 32.8K | 118.9K |
Benchmarks
Higher is better. Bold marks the winner where both labs report a score. ≈ marks the index model's estimate for a score that wasn't reported, ✓ an independent measurement and * a score measured by another lab.
| Benchmark | LFM2.5 24B-A2B | Mistral Small 3.2 |
|---|---|---|
| MMLU-Pro Knowledge | ≈ 79 | 69.1 |
| GPQA Diamond Knowledge | 74.4 | 46.1 |
| Humanity's Last Exam Reasoning | 7.4 | ≈ 4 |
| SWE-bench Verified Coding | 51.0 | ≈ 19 |
| IFBench Instruction following | 74.7 | — |
| IFEval Instruction following | 96.8 | — |
| BFCL Agentic | 65.0 | — |
| MATH-500 Math | 96.0 | — |
| SciCode Coding | 32.7 | — |
Scores are mostly reported by each lab with its own settings, so small gaps are noise. See the methodology. Open in the interactive comparison to add more models.
Questions
Which is better, LFM2.5 24B-A2B or Mistral Small 3.2?
LFM2.5 24B-A2B scores higher on the Small LM Index: 61.4 vs 38.6. On the 1 benchmark both labs report, LFM2.5 24B-A2B wins 1 and Mistral Small 3.2 wins 0.
Which is cheaper to run?
Mistral Small 3.2 is cheaper through an API: $0.11 per million tokens against $0.21. Both need about 16.9 GB of memory at 4-bit quantization.
Which has the longer context window?
Both have a 128K-token context window.
Can I use LFM2.5 24B-A2B and Mistral Small 3.2 commercially?
LFM2.5 24B-A2B is released under LFM Open License and Mistral Small 3.2 under Apache-2.0. Apache 2.0 and MIT allow commercial use; other licenses can add conditions, so read the license before shipping a product.