LFM2.5-8B-A1B

Home/Models/unsloth/LFM2.5-8B-A1B

unsloth/LFM2.5-8B-A1B

kind: original
quant: bf16 (16-bit)
family: misc
modality: text-generation
root model: LiquidAI/LFM2.5-8B-A1B
provider: unsloth
license: other
language: en, ar, zh, fr, de, ja, ko, es, pt
context: 131,072
library: transformers
vRAM (GB): —
size on disk: —
downloads: 1,200
likes: 13
created: —
last modified: 2026-05-30T10:13:04+00:00 (2026)
huggingface: https://huggingface.co/unsloth/LFM2.5-8B-A1B
tags: transformerssafetensorslfm2_moetext-generationliquidunslothlfm2.5edgeconversationalenarzhfrdejakoesptarxiv:2511.23404base_model:LiquidAI/LFM2.5-8B-A1Bbase_model:finetune:LiquidAI/LFM2.5-8B-A1Blicense:othereval-resultsendpoints_compatible

benchmark	category	score	context	conf	evidence
IFEval	instruction-following	91.84		high	`\| LFM2.5-8B-A1B \| 8B/A1B \| -24.70 \| 8.67 \| 63.47 \| 91.84 \| 56.47 \| 79.93 \| \|`
MATH-500	math	88.76		high	`\| LFM2.5-8B-A1B \| 8B/A1B \| 88.76 \| 42.53 \| 50.00 \| 64.79 \| 49.73 \| 88.07 \| 39.82 \|`
IFEval	instruction-following	79.44	IFEval	medium	`\| IFEval \| 79.44 \| 91.84 \| +12.40 \|`
MATH-500	math	74.8	MATH500	medium	`\| MATH500 \| 74.80 \| 88.76 \| +13.96 \|`
IFBench	instruction-following	56.47		high	`\| LFM2.5-8B-A1B \| 8B/A1B \| -24.70 \| 8.67 \| 63.47 \| 91.84 \| 56.47 \| 79.93 \| \|`
AIME 2026	math	50.0		high	`\| LFM2.5-8B-A1B \| 8B/A1B \| 88.76 \| 42.53 \| 50.00 \| 64.79 \| 49.73 \| 88.07 \| 39.82 \|`
AIME 2025	math	42.53		high	`\| LFM2.5-8B-A1B \| 8B/A1B \| 88.76 \| 42.53 \| 50.00 \| 64.79 \| 49.73 \| 88.07 \| 39.82 \|`
IFBench	instruction-following	26.0	IFBench	medium	`\| IFBench \| 26.00 \| 56.47 \| +30.47 \|`
AIME 2025	math	20.0	AIME25	medium	`\| AIME25 \| 20.00 \| 42.53 \| +22.53 \|`

none scraped.