LLM Reference

LFM2.5 1.2B Instruct

Released
2026-02-01
Last refreshed
2026-05-19
Status
Researched 23d ago
ProprietaryCommercial use with conditionsAgentsClassificationJSON / Tool use

LFM2.5 1.2B Instruct has model metadata, but missing tracked provider pricing keeps it from being a default production pick.

Use it for

  • Teams evaluating agents, classification, and json / tool use
  • Workloads that can use a 32k 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
Specifications
Family
LFM-2.5
Released
2026-02-01
Context
32k
Parameters
1.2B
Architecture
Decoder Only
Knowledge cutoff
2024
Specialization
general
Openness
Proprietary
License
ProprietaryCommercial use with conditions
Training
pretrained
Created by

Adaptive and efficient AI models

Cambridge, MA, United States
Founded 2023
Website
Pricing

No tracked provider token pricing is available yet.

About

LFM2.5-1.2B-Instruct is a compact, high-performance instruction-tuned model from LiquidAI built for fast on-device AI. It delivers strong results on benchmarks relative to its size.

LFM2.5 1.2B Instruct is a proprietary model in the LFM-2.5 family. The structured metadata tracks a 32k-token context window, function calling, and tool use. Headline tracked benchmarks include AIME 2025 14.0, Berkeley Function Calling Leaderboard v3 49.1, and Google-Proof Q&A 38.9.

Top use-case fit: coding, agents, and build tasks

Agents

1 relevant benchmark in the decision map.

Classification

1 relevant benchmark in the decision map.

JSON / Tool use

1 relevant benchmark in the decision map.

Provider price ladder

No tracked provider token pricing is available for this model yet.

Capabilities

Function CallingTool Use

Benchmark peer barsfor Agents

Benchmark scores(4)

Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.
BenchmarkScoreVersionSource
AIME 202514.0AIME 2025 (accuracy)https://huggingface.co/LiquidAI/LFM2.5-1.2B-Instruct
Berkeley Function Calling Leaderboard v349.1BFCLv3 (accuracy)https://huggingface.co/LiquidAI/LFM2.5-1.2B-Instruct
Google-Proof Q&A38.9GPQA (accuracy)https://huggingface.co/LiquidAI/LFM2.5-1.2B-Instruct
MMLU PRO44.4From official model card (accuracy)https://huggingface.co/LiquidAI/LFM2.5-1.2B-Instruct

Migration checks

No linked migration route is available for this model yet.