LFM2 2.6B
Sep 2025
LFM2 2.6BCompare inon
Benchmarks of LFM2 2.6B13
BenchmarkByScoreRank
Overall1
Knowledge4
Science2
Coding1
Documents1
Legacy · saturated or no longer reported4
No benchmark matches
—per task?
self-host onlyprice
—output speed?
33Kcontext
Ranking?
Alternatives?
ModelScoreCostPer taskWhat changes
BetterGranite 4.2 3BthinkingOpen+83%—$0.006Memory (4-bit): 1.3 GB → 1.8 GBNOpenNanbeige4.1 3B×2.8—$0.10 / $0.20Memory (4-bit): 1.3 GB → 2 GB
BetterMinistral 3 3B 2512Open+17%—$0.008Memory (4-bit): 1.3 GB → 1.9 GB
BetterGemma 4 E4BthinkingOpen+88%—$0.008Memory (4-bit): 1.3 GB → 4 GB
BetterGPT-5.6 Lunalow×2.7—$0.01ModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
Granite 4.2 3BthinkingOpen+83%—$0.006Memory (4-bit): 1.3 GB → 1.8 GBMinistral 3 3B 2512Open+17%—$0.008Memory (4-bit): 1.3 GB → 1.9 GBGemma 4 E4BthinkingOpen+88%—$0.008Memory (4-bit): 1.3 GB → 4 GBGPT-5.6 Lunalow×2.7—$0.01GPT OSS 20BhighOpen×2.2—$0.011Memory (4-bit): 1.3 GB → 13 GBModelScoreCostPer taskWhat changes
NNanbeige4.1 3BOpen×2.8—$0.10 / $0.20Memory (4-bit): 1.3 GB → 2 GB
Qwen 3.5 9BthinkingOpen×2.6—$0.21Memory (4-bit): 1.3 GB → 4.8 GBQwen3.5 4BthinkingOpen×2.5—$0.03 / $0.15Memory (4-bit): 1.3 GB → 2.3 GBGemma 4 12BthinkingOpen×2.5—$0.10 / $0.30Memory (4-bit): 1.3 GB → 6 GBIK2 Horizon 7BthinkingOpen×2.5—$0.30 / $1.20Memory (4-bit): 1.3 GB → 4.5 GBModelScoreCostPer taskWhat changes
LFM2.5 8B A1BthinkingOpen+68%——Memory (4-bit): 1.3 GB → 4.2 GBLFM2 24B A2Bnon-thinkingOpen+55%—$0.03 / $0.12Memory (4-bit): 1.3 GB → 14 GBLFM2 8B A1BOpen+13%——Memory (4-bit): 1.3 GB → 4.2 GBLFM2.5 1.2B ThinkingthinkingOpen+11%——LFM2.5 1.2B Instructnon-thinkingOpen+7%——LFM2.5 VL 1.6Bnon-thinkingOpen-6%——LFM2 1.2BOpen-25%——All Liquid AI models →Run it yourself
estimateFits on✓ 8 GB (phone, laptop) · full quality✓ 16 GB · full quality✓ 24 GB GPU · full quality✓ 48 GB GPU · full quality✓ 80 GB GPU · full quality
Licencelfm1.0commercial use with conditions
Benchmarks
all categoriesChat & Knowledge
Humanity's Last Exam5.5#263
AA Omniscience Index-54.1#234
CritPt (physics)0#182
AA Intelligence Index 4.3.2≈5.2#318
AA Omniscience · Accuracy5.4#315
GPQA Diamond30.6#297
AIME 2025 · AA archived8.3#128
Coding
LiveCodeBench · AA archived8.1#147
Agents & Tool Use
τ²-bench · AA13.5#214
Writing & Language
IFBench (instructions)19.5#245
FAQ
How much does LFM2 2.6B cost?›
There is no per-unit API price: self-host only.
Is LFM2 2.6B free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is LFM2 2.6B, and what is it best at?›
#261 of 336 in Chat & Knowledge, #203 of 239 in Coding, #211 of 242 in Agents & Tool Use, #242 of 245 in Writing & Language.
Is LFM2 2.6B better than Granite 4.1 3B?›
On GPQA Diamond, LFM2 2.6B scores 30.6 vs 31.4 — about the same.
What is the context window of LFM2 2.6B?›
33K tokens.
When was LFM2 2.6B released? Is there a newer version?›
Released 2025-09-23. It has been superseded by LFM2.5 2.6B.
Is LFM2 2.6B open source?›
Yes, the weights are open (lfm1.0 licence).
Can I run LFM2 2.6B locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can LFM2 2.6B read images?›
No, it accepts text only.
What is LFM2 2.6B?›
LFM2 2.6B is an open-weights language model from Liquid AI, released Sep 2025.
What are the best alternatives to LFM2 2.6B?›
Granite 4.2 3B (better), Nanbeige4.1 3B (open weights).