LFM2.5-2.6B (free)
Liquid · released Aug 11, 2026
LFM2.5-2.6B is a compact reasoning model from Liquid AI. It is suited for agent workflows, data extraction, RAG, and long-context processing. Liquid advises against using it for agentic coding or...
Specification
- Context window
- 66K
- Max output
- 8K
- Knowledge cutoff
- Not stated
- Parameters
- 2.7B
- Licence
- other
- Serving providers
- 1
- Moderated
- No
- Uptime
- 100.0%
Intelligence
8.4
33th percentile
Coding
7.7
Coding Index
Agentic
—
Agentic Index
Output speed
203 t/s
Median across providers
Latency
303ms
Time to first token
Cost per task
Free
Estimated
Benchmarks
Where the score comes from
The Intelligence Index is a composite. These are the underlying evaluations this model was actually measured on.
Evaluation scores
Percentage correct · higher is better
- GPQA Diamond55.8%
- SciCode14.4%
- τ²-bench (Telecom)7.2%
- Humanity's Last Exam6.2%
- AA-LCR (long context)5.7%
An evaluation missing from this list was not run for this model — it is not a zero.
View as table
| Evaluation | Score |
|---|---|
| GPQA Diamond | 55.8% |
| SciCode | 14.4% |
| τ²-bench (Telecom) | 7.2% |
| Humanity's Last Exam | 6.2% |
| AA-LCR (long context) | 5.7% |
Against its peers
Intelligence Index · this model highlighted, nearest peers in grey
- gpt-oss-20bOpenAI9.0
- Mistral Medium 3Mistral AI9.0
- LFM2.5-2.6B (free)Liquid8.4
- GLM 4.6VZ.ai8.4
- GPT-4o (2024-11-20)OpenAI8.4
- GPT-4oOpenAI8.4
- 7.9
- Solar Pro 3Upstage7.8
Peers are the models sitting closest on the Intelligence Index — the set you would realistically choose between.
View as table
| Model | Intelligence |
|---|---|
| gpt-oss-20b | 9.0 |
| Mistral Medium 3 | 9.0 |
| LFM2.5-2.6B (free) | 8.4 |
| GLM 4.6V | 8.4 |
| GPT-4o (2024-11-20) | 8.4 |
| GPT-4o | 8.4 |
| Qwen3 VL 30B A3B Instruct | 7.9 |
| Solar Pro 3 | 7.8 |
Percentile among all indexed models
Pricing
What it costs to run
List prices per million tokens, plus what one representative task works out to.
List price
- Input / 1M tokens
- Free
- Output / 1M tokens
- Free
- Cached input / 1M
- Not offered
- Blended 3:1
- Free
One task, estimated
Free
- Input tokens
- 50,000
- Output tokens
- 80,000
- Profile
- Reasoning
Estimated from list pricing: 50K input tokens plus 80K output tokens for reasoning models (25K for non-reasoning).
Serving providers
Speed and latency figures are medians across these providers, so a widely-served model reports a blend rather than any single endpoint.