Compare / LFM2.5 2.6B vs Llama 3.2 3B
LFM2.5 2.6B vs Llama 3.2 3B
Updated November 2026
Verdict
LFM2.5 2.6B scores higher on the Small LM Index: 44.7 vs 19.1. On the 6 benchmarks both labs report, LFM2.5 2.6B wins 6 and Llama 3.2 3B wins 0. Both need about 3.4 GB of memory at 4-bit quantization. Llama 3.2 3B is cheaper through an API: $0.12 per million tokens against $0.39.
Pick LFM2.5 2.6B if…
- you want the strongest model overall
- you mostly write code
- you need math and reasoning
- you need precise instruction following
Pick Llama 3.2 3B if…
- API cost matters most
- you want the more widely used model (more quantized builds and community support)
Specifications
| LFM2.5 2.6B | Llama 3.2 3B | |
|---|---|---|
| Small LM Index | 44.7 (likely 42–48) | 19.1 (likely 18–21) |
| Lab | Liquid AI · United States | Meta · United States |
| Released | 2026-07-28 | 2024-09-25 |
| Parameters | 2.7B (dense) | 3.2B (dense) |
| Memory at 4-bit | ~2.3 GB | ~3.4 GB |
| Context window | 128K tokens | 128K tokens |
| Input | text | text |
| Reasoning | No thinking | No thinking |
| License | LFM Open License | Llama 3.2 Community |
| API price / 1M tokens | $0.39 ($0.35 in, $0.50 out) | $0.12 ($0.050 in, $0.33 out) |
| Hugging Face downloads (30 days) | 67.9K | 966K |
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 2.6B | Llama 3.2 3B |
|---|---|---|
| MMLU-Pro Knowledge | ≈ 67 | 34.7✓ |
| GPQA Diamond Knowledge | 55.8✓ | 25.5✓ |
| Humanity's Last Exam Reasoning | 6.2✓ | 5.3✓ |
| AIME Math | 51.9 | 6.7✓ |
| LiveCodeBench Coding | 59.4 | 8.3✓ |
| SWE-bench Verified Coding | 18.0 | ≈ 29 |
| τ²-Bench Agentic | — | 21.1✓ |
| IFBench Instruction following | 59.2 | 26.2✓ |
| IFEval Instruction following | 95.4 | — |
| BFCL Agentic | 56.9 | — |
| MATH-500 Math | — | 48.9✓ |
| AA-LCR Long context | 5.7✓ | 4.3✓ |
| SciCode Coding | 14.4✓ | — |
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 2.6B or Llama 3.2 3B?
LFM2.5 2.6B scores higher on the Small LM Index: 44.7 vs 19.1. On the 6 benchmarks both labs report, LFM2.5 2.6B wins 6 and Llama 3.2 3B wins 0.
Which is cheaper to run?
Llama 3.2 3B is cheaper through an API: $0.12 per million tokens against $0.39. Both need about 3.4 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 2.6B and Llama 3.2 3B commercially?
LFM2.5 2.6B is released under LFM Open License and Llama 3.2 3B under Llama 3.2 Community. Apache 2.0 and MIT allow commercial use; other licenses can add conditions, so read the license before shipping a product.