LFM2.5-2.6B
Liquid AIOpen weightsReasoningProvisionalReleased Aug 4, 2026Input: text
Overall score
31.2
Rank 178 of 222 · 16.9–45.6
Context window
128K
128,000 tokens
Price per 1M tokens
— in · — out
No price listed
Blended price
—
3:1 input:output tokens
Output speed
—
Not measured yet
Category scores
Scores per capability area with the model's rank among ranked models in that category.
- 1Instruction following#11939.1
- 2Knowledge30.7
- 3Reasoning28.9
Benchmark results (7)
As reported in system cards, papers and public evaluations. Position compares this model with every other model reporting the same benchmark.
Agentic & tool use
| Claw-Eval | 62.9 | #17 of 36 |
| τ³-bench results | 5.7 | #23 of 23 |
| BFCL v4 | 56.9 | #11 of 19 |
| PinchBench | 68.2 | #7 of 7 |
Coding
| LiveCodeBench v6 | 59.4 | #38 of 46 |
Instruction following
| IFBench | 59.2 | #32 of 42 |
Math
| AIME 2025 | 51.9 | #28 of 38 |
Pricing by source
USD per 1M tokens.
| Source | Input | Output | Cached in |
|---|---|---|---|
| List price | — | — | — |
| OpenRouter liquid/lfm-2.5-2.6b:free | — | — | — |
Liquid AI's August 4, 2026 launch post and Hugging Face model card publish LFM2.5-2.6B under the LFM Open License v1.0 for local, on-device, and self-hosted use. Liquid does not publish a separate hosted API token price for this checkpoint, so BenchLM represents it as self-host/free-per-token before infrastructure costs.
Open weights
From the Hugging Face repository.
- Parameters
- 2.7B
- Licence
- other
- Downloads (30 days)
- 103.5K
- Repository
- LiquidAI/LFM2.5-2.6B ↗
Speed
Measured by us through OpenRouter (median of trials).
Not measured yet.
Related models
Same family
Similar score
- Gemma 4 E4B31.1 · —
- Sarvam 105B31.7 · Free
- DeepSeek R1 Distill Qwen 32B30.8 · —
- Ornith-1.5-9B31.7 · —
- DeepSeek V331.9 · $0.48