LFM2 24B A2B
LFM2 24B A2BRankingin non-thinking mode?
Alternatives?
CheaperNemotron 3.5 LightningthinkingOpen×2.2-7%$0.079Memory (4-bit): 14 GB → 20 GB
BetterGPT-6 Lunalow×3.6+281%$0.005Function calling only with none · Chat Completions+3
OpenQwen3.8 27Bxhigh×5.7+1,721%$1.01Memory (4-bit): 14 GB → 17 GB+5CheaperGPT OSS 20BlowOpen+67%-31%$0.018 / $0.09
CheaperGemma 4 E4BthinkingOpen+50%-24%$0.008Nemotron 3.5 LightningthinkingOpen×2.2-7%$0.079Memory (4-bit): 14 GB → 20 GBGPT OSS 20BlowOpen+67%-31%$0.018 / $0.09Gemma 4 E4BthinkingOpen+50%-24%$0.008Qwen3.5 2BthinkingOpensame-24%$0.02 / $0.10Qwen3.5 0.8BthinkingOpensame-62%$0.01 / $0.05GPT-6 Lunalow×3.6+281%$0.005Function calling only with none · Chat Completions+3
Granite 4.2 3BthinkingOpen+52%same$0.006Gemma 4 E4BthinkingOpen+50%-24%$0.008GPT-5.6 Lunalow×3.5+757%$0.01GPT OSS 20BhighOpen+51%-31%$0.011Qwen3.8 27BxhighOpen×5.7+1,721%$1.01Memory (4-bit): 14 GB → 17 GB+5IK2 Horizon MoVA 36B A4BthinkingOpen×4.3+3,233%$1 / $4Memory (4-bit): 14 GB → 22 GBQwen 3.5 27BthinkingOpen×3.9+1,471%$0.30 / $2.40Memory (4-bit): 14 GB → 17 GBAG9v3 39A5BOpen×3.7——Memory (4-bit): 14 GB → 23 GBQwen3.6 27BthinkingOpen×3.6+2,471%$0.62Memory (4-bit): 14 GB → 17 GBRun it yourself
estimateBenchmarks
all categoriesFAQ
How much does LFM2 24B A2B cost?›
$0.03 per 1M input tokens and $0.12 per 1M output tokens (cheapest host: Together AI).
Is LFM2 24B A2B free to use?›
There is no free API; the cheapest paid price is $0.03 per 1M input tokens. The weights are open, so you can run it yourself.
How good is LFM2 24B A2B, and what is it best at?›
#300 of 334 in Chat & Knowledge, #210 of 239 in Coding, #222 of 242 in Agents & Tool Use, #127 of 245 in Writing & Language.
Is LFM2 24B A2B better than Qwen3 8B?›
On AA Intelligence Index 4.3.2, LFM2 24B A2B scores 5.9 vs 6 — about the same.
How can I use LFM2 24B A2B?›
Through the Liquid AI API.
What is the context window of LFM2 24B A2B?›
33K tokens.
When was LFM2 24B A2B released? Is there a newer version?›
Released 2026-02-25. It is the current version in its line.
Is LFM2 24B A2B open source?›
Yes, the weights are open (lfm1.0 licence).
Can I run LFM2 24B A2B locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can LFM2 24B A2B read images?›
No, it accepts text only.
What is LFM2 24B A2B?›
LFM2 24B A2B is an open-weights language model from Liquid AI, released Feb 2026.
What are the best alternatives to LFM2 24B A2B?›
Nemotron 3.5 Lightning (cheaper), GPT-6 Luna (better), Qwen3.8 27B (open weights).