LFM2-8B-A1B

Home/Models/unsloth/LFM2-8B-A1B

unsloth/LFM2-8B-A1B

kind: original
quant: bf16 (16-bit)
family: misc
modality: text-generation
root model: LiquidAI/LFM2-8B-A1B
provider: unsloth
license: other
language: en, ar, zh, fr, de, ja, ko, es
context: 32,768
library: transformers
vRAM (GB): —
size on disk: —
downloads: 122
likes: 12
created: —
last modified: 2025-10-08T08:46:13+00:00 (2025)
huggingface: https://huggingface.co/unsloth/LFM2-8B-A1B
tags: transformerssafetensorslfm2_moetext-generationliquidunslothlfm2edgemoeconversationalcustom_codeenarzhfrdejakoesbase_model:LiquidAI/LFM2-8B-A1Bbase_model:finetune:LiquidAI/LFM2-8B-A1Blicense:otherendpoints_compatibleregion:us

benchmark	category	score	conf	evidence
GSM8K	math	84.38	low	`\| LFM2-8B-A1B \| 84.38 \| 64.76 \| 74.2 \| 62.38 \| 72.4 \| 55.26 \|`
IFEval	instruction-following	77.58	low	`\| LFM2-8B-A1B \| 64.84 \| 37.42 \| 29.29 \| 77.58 \| 25.85 \| 58.19 \|`
MATH-500	math	74.2	low	`\| LFM2-8B-A1B \| 84.38 \| 64.76 \| 74.2 \| 62.38 \| 72.4 \| 55.26 \|`
MGSM	math	72.4	low	`\| LFM2-8B-A1B \| 84.38 \| 64.76 \| 74.2 \| 62.38 \| 72.4 \| 55.26 \|`
HumanEval	coding	69.51	low	`\| LFM2-8B-A1B \| 1.5B \| 21.04% \| 21.36% \| 69.51% \| 44.22% \|`
MMLU	knowledge	64.84	low	`\| LFM2-8B-A1B \| 64.84 \| 37.42 \| 29.29 \| 77.58 \| 25.85 \| 58.19 \|`
MATH	math	62.38	low	`\| LFM2-8B-A1B \| 84.38 \| 64.76 \| 74.2 \| 62.38 \| 72.4 \| 55.26 \|`
MMLU-Pro	knowledge	37.42	low	`\| LFM2-8B-A1B \| 64.84 \| 37.42 \| 29.29 \| 77.58 \| 25.85 \| 58.19 \|`
GPQA	reasoning	29.29	low	`\| LFM2-8B-A1B \| 64.84 \| 37.42 \| 29.29 \| 77.58 \| 25.85 \| 58.19 \|`
IFBench	instruction-following	25.85	low	`\| LFM2-8B-A1B \| 64.84 \| 37.42 \| 29.29 \| 77.58 \| 25.85 \| 58.19 \|`

none scraped.