LFM2 8B A1B
Oct 2025
LFM2 8B A1BCompare inon
Benchmarks of LFM2 8B A1B12
BenchmarkByScoreRank
Overall1
Knowledge4
Science2
Coding1
Legacy · saturated or no longer reported4
No benchmark matches
—per task?
self-host onlyprice
—output speed?
33Kcontext
Ranking?
Alternatives?
ModelScoreCostPer taskWhat changes
BetterGPT-6 Lunalow×4.4—$0.005Function calling only with none · Chat Completions+3IOpenK2 Horizon 7Bthinking×4.2—$0.30 / $1.20Memory (4-bit): 4.2 GB → 4.5 GB
BetterGranite 4.2 3BthinkingOpen+84%—$0.006
BetterGemma 4 E4BthinkingOpen+81%—$0.008BetterGPT-5.6 Lunalow×4.3—$0.01ModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
GPT-6 Lunalow×4.4—$0.005Function calling only with none · Chat Completions+3Granite 4.2 3BthinkingOpen+84%—$0.006Gemma 4 E4BthinkingOpen+81%—$0.008GPT-5.6 Lunalow×4.3—$0.01GPT OSS 20BhighOpen+82%—$0.011Memory (4-bit): 4.2 GB → 13 GBModelScoreCostPer taskWhat changes
IK2 Horizon 7BthinkingOpen×4.2—$0.30 / $1.20Memory (4-bit): 4.2 GB → 4.5 GBIK2 Horizon 3.7BthinkingOpen×3.2—$0.30 / $1.20Gemma 4 12BthinkingOpen×2.9—$0.10 / $0.30Memory (4-bit): 4.2 GB → 6 GBSApriel v1.5 15B ThinkerOpen×2.8——Memory (4-bit): 4.2 GB → 7.4 GBSApriel v1.6 15B ThinkerOpen×2.7——Memory (4-bit): 4.2 GB → 7.4 GBModelScoreCostPer taskWhat changes
LFM2.5 8B A1BthinkingOpen+46%——LFM2 24B A2Bnon-thinkingOpen+21%—$0.03 / $0.12Memory (4-bit): 4.2 GB → 14 GBLFM2.5 1.2B ThinkingthinkingOpensame——LFM2 2.6BOpensame——LFM2.5 1.2B Instructnon-thinkingOpensame——LFM2.5 VL 1.6Bnon-thinkingOpensame——LFM2 1.2BOpensame——All Liquid AI models →Run it yourself
estimateFits on✓ 8 GB (phone, laptop) · compressed✓ 16 GB · 8-bit✓ 24 GB GPU · full quality✓ 48 GB GPU · full quality✓ 80 GB GPU · full quality
Licencelfm1.0commercial use with conditions
History
2 versionsBenchmarks
all categoriesChat & Knowledge
Humanity's Last Exam4.9#279
AA Omniscience Index-76.5#305
CritPt (physics)0#180
AA Intelligence Index 4.3.2≈4.9#325
AA Omniscience · Accuracy8.2#295
GPQA Diamond34.4#289
AIME 2025 · AA archived25.3#111
Coding
LiveCodeBench · AA archived15.1#140
Agents & Tool Use
τ²-bench · AA10.5#227
Writing & Language
IFBench (instructions)26.3#227
FAQ
How much does LFM2 8B A1B cost?›
There is no per-unit API price: self-host only.
Is LFM2 8B A1B free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is LFM2 8B A1B, and what is it best at?›
#277 of 334 in Chat & Knowledge, #210 of 239 in Coding, #224 of 242 in Agents & Tool Use, #224 of 245 in Writing & Language.
Is LFM2 8B A1B better than Molmo2 8B?›
On AA Intelligence Index 4.3.2, LFM2 8B A1B scores 4.9 vs 5 — about the same.
What is the context window of LFM2 8B A1B?›
33K tokens.
When was LFM2 8B A1B released? Is there a newer version?›
Released 2025-10-07. It has been superseded by LFM2.5 8B A1B.
Is LFM2 8B A1B open source?›
Yes, the weights are open (lfm1.0 licence).
Can I run LFM2 8B A1B locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can LFM2 8B A1B read images?›
No, it accepts text only.
What is LFM2 8B A1B?›
LFM2 8B A1B is an open-weights language model from Liquid AI, released Oct 2025.
What are the best alternatives to LFM2 8B A1B?›
GPT-6 Luna (better), K2 Horizon 7B (open weights).