LFM3-1.7B
A compact 1.7B-parameter text model with a 256K-token context, post-trained for agents, tool calling, and structured outputs. It reasons before it answers.
Specifications
| Lab | Liquid AI |
|---|---|
| Context window | 262.1K tokens |
| Input price | $0.08/1M |
| Output price | $0.20/1M |
| Release | Nov 2026 |
Strengths
- Compact 1.7B footprint suited to edge and resource-constrained agent deployments
- 256K-token context window, large for a model this size
- Post-trained for tool calling and structured outputs
- Reasons before answering, with structured reasoning support
- Open-weight release available on Hugging Face
Weaknesses
- Small parameter count caps performance on complex reasoning versus larger models
- New release with limited independent benchmark coverage so far
- Via OpenRouter, prompts and outputs may be retained and used for training
Best for
- Lightweight AI agents
- Tool-calling and structured-output pipelines
- Cost-sensitive high-volume inference
- Edge or resource-constrained deployments
In Depth: LFM3-1.7B
Summary
LFM3-1.7B is an AI model from Liquid AI.
Released in Nov 2026. It takes text input and produces text output, with a context window of 262.1K tokens. Input pricing is $0.08 per 1M tokens and output is $0.20 per 1M tokens on OpenRouter.