LFM2.5 8B A1B
Last refreshed 2026-05-28. Next refresh: weekly.
LFM2.5 8B A1B has model metadata, but missing tracked provider pricing keeps it from being a default production pick.
Decision context: RAG task fit, 0 tracked provider routes, and research from 2026-05-28.
Use it for
- Teams evaluating rag, agents, and long context
- Workloads that can use a 128k context window
Do not use it for
- Cost-sensitive launches that need sourced token pricing
- Vision or document-understanding workloads
- Teams that need a tracked hosted API route today
Cheapest output
-
No tracked output price
Provider routes
0
No provider route in seed
Quality / dollar
Unknown
No task benchmark coverage yet
Freshness
2026-05-28
Researched 1d ago
Top use-case fit
RAG
Included by capability and metadata signals in the decision map.
Agents
Included by capability and metadata signals in the decision map.
Long context
Included by capability and metadata signals in the decision map.
Provider price ladder
No tracked provider token pricing is available for this model yet.
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.
About
LFM2.5-8B-A1B is Liquid AI's latest on-device mixture-of-experts model, succeeding LFM2-8B-A1B. It has 8.3B total parameters with approximately 1.5B active per token (the A1B label uses a rounded ~1B figure). The architecture combines 18 double-gated LIV convolutional layers with 6 GQA attention layers, trained on 38 trillion tokens. The context window expands to 128K tokens (up from 32K in the predecessor). It is a reasoning model that generates explicit chain-of-thought steps before producing its final answer, making reasoning tokens cheap due to the MoE design. Strong tool-calling, function-calling, and instruction-following capabilities make it well-suited for agentic workflows on edge hardware. Weights are openly available on Hugging Face under the lfm1.0 license.
LFM2.5 8B A1B has a 128k-token context window.
Capabilities
Benchmark Scores(3)
| Benchmark | Score | Version | Source |
|---|---|---|---|
| Instruction-Following Evaluation | 91.8 | — | https://www.liquid.ai/blog/lfm2-5-8b-a1b |
| MATH-500 | 88.8 | — | https://www.liquid.ai/blog/lfm2-5-8b-a1b |
| AIME 2025 | 42.5 | — | https://www.liquid.ai/blog/lfm2-5-8b-a1b |
