LFM2.5-2.6B
LFM2.5-2.6B is available now for rag, agents, and long context with open-weight and 131k context; evaluate it while provider pricing coverage matures.
- Family
- LFM-2.5
- Released
- 2026-08-04
- Context
- 131k
- Parameters
- 2.69B
- Architecture
- Decoder Only
- Specialization
- general
- Openness
- Open weights
- License
- LFM Open License v1.0Commercial use: conditional
- Weights
- Available
- Code
- Unknown
- Training
- Multi-stage
No tracked provider token pricing is available yet.
About
Liquid AI LFM2.5-2.6B is a 2.69B-parameter hybrid (LFM2 architecture) on-device model with 131,072-token context and agentic post-training. It is a pure reasoning model that always emits a <think> block, and it writes Pythonic function calls between tool-call tokens. It covers 16 languages. Liquid recommends it for agents, tool use, data extraction, RAG and long-context work, but not for agentic coding or knowledge-heavy tasks. Released 4 August 2026 under the LFM Open License v1.0, with GGUF and DSpark variants.
Provider price ladder
No tracked provider token pricing is available for this model yet.
Capabilities
Benchmark peer barsfor RAG
No task-mapped benchmark peers are available for this model yet.
Migration checks
No linked migration route is available for this model yet.
API versions
LiquidAI/LFM2.5-2.6BNo tracked provider token pricing is available yet.