liquid
LiquidAI: LFM2.5-2.6B (free)
liquid/lfm-2.5-2.6b:free
Our take
AI-assisted analysis based on the sources below. Suitability is an editorial assessment; test it on your own workload.
liquid/lfm-2.5-2.6b:free is the OpenRouter free listing for Liquid AI's LFM2.5-2.6B, a compact text-only reasoning model documented for on-device and local deployment. Liquid AI describes post-training for agentic work, function calling, data extraction, retrieval-augmented generation, and long-context processing, while the catalog marks reasoning as mandatory. Vendor-reported results emphasize efficient CPU, Apple Silicon, and high-concurrency GPU inference; these are measured claims rather than guarantees. The same documentation advises against agentic coding and knowledge-heavy tasks.[1][2]
Strengths
- Compact, text-only model documented for on-device deployment and supported across Transformers, vLLM, SGLang, llama.cpp, MLX, and LM Studio.[2]
- The documentation describes agentic post-training and a four-step function-calling workflow, with tools defined through the chat template.[2]
- It is positioned for long-context processing, RAG, and data extraction; vendor tests report fast CPU and Apple Silicon decoding plus high-concurrency GPU serving.[1][2]
Limitations
- Liquid AI explicitly advises against using the model for agentic coding and knowledge-heavy tasks.[2]
- Reasoning is not optional in this variant: the catalog marks it as mandatory and the model card says every answer begins with a thinking segment.[1][2]
- The sources disagree on the advertised context length and no independent evaluation is supplied; validate context, quality, and performance on the intended workload.[1][2]
Where to use it
- Local or edge assistants and agents where a compact text model and documented tool calls fit the product constraints.[2]
- Data-extraction and retrieval-augmented workflows using long context, after checking accuracy and context behavior on real documents.[1][2]
- Multilingual agent prototypes for the languages listed by the vendor, without assuming uniform quality that has not been independently measured.[2]
Sources
- [1] OpenRouter catalog · LiquidAI: LFM2.5-2.6B (free)Sources checked: Nov 13, 2026
- [2] Model documentation · LiquidAI: LFM2.5-2.6B (free)Sources checked: Nov 13, 2026
About this model
Description below is the original text supplied by OpenRouter, shown in the source language.
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...
Specifications
- Model ID
- liquid/lfm-2.5-2.6b:free
- Canonical ID
- liquid/lfm-2.5-2.6b-20260811
- Creator
- liquid
- Added to OpenRouter
- 11.08.2026
- Context length
- 65,536
- Modality
- text->text
- Input modalities
- text
- Output modalities
- text
- Token pricing
- USD per 1M tokens: in $0 / out $0
- Knowledge cutoff
- —
- Expiration date
- 13.11.2026
Tasks and capabilities
A grade appears only where the source supplies a matching measurement for that task, the model is active and the cohort holds at least 10 comparable models. The capability column lists facts the source declares — supported formats, reasoning and tool parameters — which are not measured quality.
| Task | Grade | Source-declared capability |
|---|---|---|
| General quality | No measurement | reasoning: mandatory · parameters: include_reasoning, reasoning |
| Coding | CC — below the B band | — |
| Agents | No measurement | parameters: structured_outputs, tool_choice, tools |
| Writing | No measurement | output: text |
| Reasoning | No measurement | reasoning: mandatory · parameters: include_reasoning, reasoning |
| Document analysis | No measurement | — |
| Vision | No measurement | — |
| Image generation | No measurement | — |
| Audio | No measurement | — |
| Video | No measurement | — |
| Embeddings | No measurement | — |
Evaluations
| Source | Benchmark | Value | Grade | Data retrieved |
|---|---|---|---|---|
| Artificial Analysis | Coding index | 7.7 | CC — below the B band | 13.11.2026 |
How grades work
Each letter grade compares a model only against other models scored by the same source and metric in the current snapshot. S means the top 10%, A the next 20%, B the next 30%, C the rest. Models without a published measurement are shown as not rated — a supported format alone never implies measured quality.
- S — S — top 10% of the cohort
- A — A — next 20% of the cohort
- B — B — next 30% of the cohort
- C — C — below the B band
Providers
Optional live provider statistics from OpenRouter. Missing values are unknown, not zero.
Provider details are unavailable right now; the catalog data above is unaffected.
Provider data retrieved: 13.11.2026
Related articles
No published articles identify this model yet.