model family

LFM2.5

Published by LiquidAI. 1 member measured on real machines, across 1 build. Every number below was signed by the machine that produced it — none of it is from a model card.

what it takes, what it returns
text → text

Declared by the network’s catalog, not measured — a signature says what this family would be asked. Whether it answered is the signed capability flags on each member.

1 of 58 builds measured · 57 still open
LFM2.5 1.2B1 build measured →
hf.co/unsloth/LFM2.5-1.2B-Instruct-GGUF:Q4_K_M1.2b · Q4_K_M · unsloth · ollama · 1 machine · 1.17B
4,096recalled
—served
16.1best tok/s
recalled 4k⌐ declared 125k · unsignedadvertises 31× what it proved

Builds this network knows of

Declared, not measured — this is what the catalog says exists and what it would cost a machine to hold. Nothing here is signed, and none of it claims the build works.

how much memory does your machine have

39 of 58 builds fit a 16 GB machine — budgeting 70% of it, or 11.2 GiB, because the cache and the runtime want the rest.

  • unsloth/LFM2.5-230MBF16vllmderived by unsloth
    0.4 GiB
  • hf.co/unsloth/LFM2.5-1.2B-Instruct-GGUF:Q2_KQ2_Kollamaquantized by unsloth
    0.5 GiB
  • hf.co/unsloth/LFM2.5-1.2B-Instruct-GGUF:Q2_K_XLQ2_K_XLollamaquantized by unsloth
    0.5 GiB
  • hf.co/unsloth/LFM2.5-1.2B-Instruct-GGUF:Q3_K_SQ3_K_Sollamaquantized by unsloth
    0.5 GiB
  • LiquidAI/LFM2.5-VL-3B-DSparkBF16vllmpublished by LiquidAI
    0.5 GiB
  • LiquidAI/LFM2.5-24B-A2B-DSparkBF16vllmpublished by LiquidAI
    0.6 GiB
  • lmstudio-community/LFM2.5-1.2B-Instruct-MLX-4bit4BITmlxquantized by lmstudio-community
    0.6 GiB
  • hf.co/unsloth/LFM2.5-1.2B-Instruct-GGUF:Q4_0Q4_0ollamaquantized by unsloth
    0.6 GiB
  • hf.co/unsloth/LFM2.5-1.2B-Instruct-GGUF:Q4_K_MQ4_K_Mollamameasured herequantized by unsloth
    0.7 GiB
  • lmstudio-community/LFM2.5-1.2B-Instruct-MLX-6bit6BITmlxquantized by lmstudio-community
    0.9 GiB
  • hf.co/unsloth/LFM2.5-VL-3B-GGUF:IQ2_MIQ2_Mollamaquantized by unsloth
    0.9 GiB
  • hf.co/unsloth/LFM2.5-VL-3B-GGUF:IQ3_XXSIQ3_XXSollamaquantized by unsloth
    1.1 GiB
  • lmstudio-community/LFM2.5-1.2B-Instruct-MLX-8bit8BITmlxquantized by lmstudio-community
    1.2 GiB
  • hf.co/unsloth/LFM2.5-1.2B-Instruct-GGUF:Q8_0Q8_0ollamaquantized by unsloth
    1.2 GiB
  • hf.co/unsloth/LFM2.5-1.2B-Instruct-GGUF:Q8_K_XLQ8_K_XLollamaquantized by unsloth
    1.3 GiB
  • mlx-community/LFM2.5-VL-1.6B-4bit4BITmlxderived by mlx-community
    1.4 GiB
  • hf.co/unsloth/LFM2.5-VL-3B-GGUF:IQ4_XSIQ4_XSollamaquantized by unsloth
    1.4 GiB
  • hf.co/unsloth/LFM2.5-VL-3B-GGUF:Q4_0Q4_0ollamaquantized by unsloth
    1.5 GiB
  • hf.co/unsloth/LFM2.5-VL-3B-GGUF:Q4_K_MQ4_K_Mollamaquantized by unsloth
    1.6 GiB
  • hf.co/DavidAU/LFM2.5-2.6B-Qwen3.8-Turbo-Brilliance-Power-X12-NEO-MAX-GGUF:IQ4_XSIQ4_XSollamaderived by DavidAU
    1.7 GiB
  • unsloth/LFM2.5-1.2B-InstructBF16vllmderived by unsloth
    2.2 GiB
  • hf.co/unsloth/LFM2.5-1.2B-Instruct-GGUF:BF16BF16ollamaquantized by unsloth
    2.2 GiB
  • hf.co/DavidAU/LFM2.5-2.6B-Qwen3.8-Turbo-Brilliance-Power-X12-NEO-MAX-GGUF:Q6_KQ6_Kollamaderived by DavidAU
    2.4 GiB
  • lmstudio-community/LFM2.5-2.6B-MLX-8bit8BITmlxquantized by lmstudio-community
    2.7 GiB
  • hf.co/unsloth/LFM2.5-VL-3B-GGUF:Q8_0Q8_0ollamaquantized by unsloth
    2.7 GiB
  • hf.co/DavidAU/LFM2.5-2.6B-Qwen3.8-Turbo-Brilliance-Power-X12-NEO-MAX-GGUF:Q8_0Q8_0ollamaderived by DavidAU
    2.9 GiB
  • hf.co/DavidAU/LFM2.5-8B-A1B-Qwen3.8-Turbo-Brilliance-Power-X12-NEO-MAX-GGUF:IQ4_XSIQ4_XSollamaderived by DavidAU
    4.6 GiB
  • Davd-b01/lfm2.5-2.6B-thinkingcap-distiller-v2BF16vllmderived by Davd-b01
    5.0 GiB
  • PhillyLabs/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASSBF16vllmderived by PhillyLabs
    5.0 GiB
  • saidutta69/lfm2.5-2.6b-fable5-coding-agent-hereticBF16vllmderived by saidutta69
    5.0 GiB
  • hf.co/DavidAU/LFM2.5-2.6B-Qwen3.8-Turbo-Brilliance-Power-X12-NEO-MAX-GGUF:BF16BF16ollamaderived by DavidAU
    5.0 GiB
  • hf.co/unsloth/LFM2.5-VL-3B-GGUF:BF16BF16ollamaquantized by unsloth
    5.0 GiB
  • LiquidAI/LFM2.5-VL-3B-MLX-bf16BF16mlxpublished by LiquidAI
    5.8 GiB
  • LiquidAI/LFM2.5-VL-3BBF16vllmpublished by LiquidAI
    5.8 GiB
  • hf.co/DavidAU/LFM2.5-8B-A1B-Qwen3.8-Turbo-Brilliance-Power-X12-NEO-MAX-GGUF:Q6_KQ6_Kollamaderived by DavidAU
    6.8 GiB
  • hf.co/unsloth/LFM2.5-24B-A2B-GGUF:IQ2_XXSIQ2_XXSollamaquantized by unsloth
    7.0 GiB
  • hf.co/unsloth/LFM2.5-24B-A2B-GGUF:IQ2_MIQ2_Mollamaquantized by unsloth
    7.0 GiB
  • hf.co/unsloth/LFM2.5-24B-A2B-GGUF:IQ3_XXSIQ3_XXSollamaquantized by unsloth
    8.3 GiB
  • hf.co/DavidAU/LFM2.5-8B-A1B-Qwen3.8-Turbo-Brilliance-Power-X12-NEO-MAX-GGUF:Q8_0Q8_0ollamaderived by DavidAU
    8.6 GiB
  • hf.co/AtomicChat/LFM2.5-24B-A2B-Uncensored-GGUF:IQ3_MIQ3_Mollamaderived by AtomicChat
    11.3 GiB
  • LiquidAI/LFM2.5-24B-A2B-MLX-4bit4BITmlxpublished by LiquidAI
    12.6 GiB
  • hf.co/LiquidAI/LFM2.5-24B-A2B-GGUF:Q4_0Q4_0ollamapublished by LiquidAI
    12.6 GiB
  • hf.co/AtomicChat/LFM2.5-24B-A2B-Uncensored-GGUF:IQ4_XSIQ4_XSollamaderived by AtomicChat
    13.5 GiB
  • hf.co/LiquidAI/LFM2.5-24B-A2B-GGUF:Q4_K_MQ4_K_Mollamapublished by LiquidAI
    13.5 GiB
  • hf.co/AtomicChat/LFM2.5-24B-A2B-Uncensored-GGUF:Q4_K_MQ4_K_Mollamaderived by AtomicChat
    14.3 GiB
  • LiquidAI/LFM2.5-24B-A2B-MLX-5bit5BITmlxpublished by LiquidAI
    15.4 GiB
  • hf.co/DavidAU/LFM2.5-8B-A1B-Qwen3.8-Turbo-Brilliance-Power-X12-NEO-MAX-GGUF:BF16BF16ollamaderived by DavidAU
    15.8 GiB
  • hf.co/AtomicChat/LFM2.5-24B-A2B-Uncensored-GGUF:Q5_K_MQ5_K_Mollamaderived by AtomicChat
    16.1 GiB
  • LiquidAI/LFM2.5-24B-A2B-FP8FP8vllmpublished by LiquidAI
    23.0 GiB
  • LiquidAI/LFM2.5-24B-A2B-MLX-8bit8BITmlxpublished by LiquidAI
    23.7 GiB
  • hf.co/LiquidAI/LFM2.5-24B-A2B-GGUF:Q8_0Q8_0ollamapublished by LiquidAI
    23.7 GiB
  • hf.co/AtomicChat/LFM2.5-24B-A2B-Uncensored-GGUF:Q8_0Q8_0ollamaderived by AtomicChat
    24.0 GiB
  • LiquidAI/LFM2.5-24B-A2BBF16vllmpublished by LiquidAI
    44.7 GiB
  • LiquidAI/LFM2.5-24B-A2B-BaseBF16vllmpublished by LiquidAI
    44.7 GiB
  • unsloth/LFM2.5-24B-A2BBF16vllmderived by unsloth
    44.7 GiB
  • LiquidAI/LFM2.5-24B-A2B-MLX-bf16BF16mlxpublished by LiquidAI
    44.7 GiB
  • hf.co/LiquidAI/LFM2.5-24B-A2B-GGUF:BF16BF16ollamapublished by LiquidAI
    44.7 GiB
  • hf.co/LiquidAI/LFM2.5-24B-A2B-GGUF:F16F16ollamapublished by LiquidAI
    44.7 GiB

Sizes are what the catalog declares, not what a machine measured — fitting on disk is necessary and not sufficient. Nobody here has run these on your machine, and this narrows the candidates rather than certifying one.

the same builds, by member
LFM2.5 1.2B
hf.co/unsloth/LFM2.5-1.2B-Instruct-GGUF:Q2_KQ2_Kollama461 MiB on diskquantized by unsloth
hf.co/unsloth/LFM2.5-1.2B-Instruct-GGUF:Q2_K_XLQ2_K_XLollama461 MiB on diskquantized by unsloth
hf.co/unsloth/LFM2.5-1.2B-Instruct-GGUF:Q3_K_SQ3_K_Sollama532 MiB on diskquantized by unsloth
lmstudio-community/LFM2.5-1.2B-Instruct-MLX-4bit4BITmlx628 MiB on diskquantized by lmstudio-community
hf.co/unsloth/LFM2.5-1.2B-Instruct-GGUF:Q4_0Q4_0ollama663 MiB on diskquantized by unsloth
hf.co/unsloth/LFM2.5-1.2B-Instruct-GGUF:Q4_K_MQ4_K_Mollama697 MiB on diskquantized by unsloth
lmstudio-community/LFM2.5-1.2B-Instruct-MLX-6bit6BITmlx907 MiB on diskquantized by lmstudio-community
lmstudio-community/LFM2.5-1.2B-Instruct-MLX-8bit8BITmlx1.2 GiB on diskquantized by lmstudio-community
hf.co/unsloth/LFM2.5-1.2B-Instruct-GGUF:Q8_0Q8_0ollama1.2 GiB on diskquantized by unsloth
hf.co/unsloth/LFM2.5-1.2B-Instruct-GGUF:Q8_K_XLQ8_K_XLollama1.3 GiB on diskquantized by unsloth
hf.co/unsloth/LFM2.5-1.2B-Instruct-GGUF:BF16BF16ollama2.2 GiB on diskquantized by unsloth
LFM2.5-VL-1.6B-4bit
mlx-community/LFM2.5-VL-1.6B-4bit4BITmlx1.4 GiB on diskderived by mlx-community
LFM2.5-1.2B-Instruct
unsloth/LFM2.5-1.2B-InstructBF16vllm2.2 GiB on diskderived by unsloth
LFM2.5 2.6B
lmstudio-community/LFM2.5-2.6B-MLX-8bit8BITmlx2.7 GiB on diskquantized by lmstudio-community
lfm2.5-2.6b-fable5-coding-agent
saidutta69/lfm2.5-2.6b-fable5-coding-agent-hereticBF16vllm5.0 GiB on diskderived by saidutta69
lfm2.5-2.6B-thinkingcap-distiller-v2
Davd-b01/lfm2.5-2.6B-thinkingcap-distiller-v2BF16vllm5.0 GiB on diskderived by Davd-b01
LFM2.5-2.6B-Qwen3.8-Turbo-Brilliance-Power-X12-NEO-MAX-GGUF
hf.co/DavidAU/LFM2.5-2.6B-Qwen3.8-Turbo-Brilliance-Power-X12-NEO-MAX-GGUF:IQ4_XSIQ4_XSollama1.7 GiB on diskderived by DavidAU
hf.co/DavidAU/LFM2.5-2.6B-Qwen3.8-Turbo-Brilliance-Power-X12-NEO-MAX-GGUF:Q6_KQ6_Kollama2.4 GiB on diskderived by DavidAU
hf.co/DavidAU/LFM2.5-2.6B-Qwen3.8-Turbo-Brilliance-Power-X12-NEO-MAX-GGUF:Q8_0Q8_0ollama2.9 GiB on diskderived by DavidAU
hf.co/DavidAU/LFM2.5-2.6B-Qwen3.8-Turbo-Brilliance-Power-X12-NEO-MAX-GGUF:BF16BF16ollama5.0 GiB on diskderived by DavidAU
LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASS
PhillyLabs/LFM2.5-2.6B-UNCENSORED-ABLITERATED-PHILADELPHIA-CLASSBF16vllm5.0 GiB on diskderived by PhillyLabs
LFM2.5-230M
unsloth/LFM2.5-230MBF16vllm438 MiB on diskderived by unsloth
LFM2.5 24B-A2B
hf.co/unsloth/LFM2.5-24B-A2B-GGUF:IQ2_XXSIQ2_XXSollama7.0 GiB on diskquantized by unsloth
hf.co/unsloth/LFM2.5-24B-A2B-GGUF:IQ2_MIQ2_Mollama7.0 GiB on diskquantized by unsloth
hf.co/unsloth/LFM2.5-24B-A2B-GGUF:IQ3_XXSIQ3_XXSollama8.3 GiB on diskquantized by unsloth
LiquidAI/LFM2.5-24B-A2B-MLX-4bit4BITmlx12.6 GiB on diskpublished by LiquidAI
hf.co/LiquidAI/LFM2.5-24B-A2B-GGUF:Q4_0Q4_0ollama12.6 GiB on diskpublished by LiquidAI
hf.co/LiquidAI/LFM2.5-24B-A2B-GGUF:Q4_K_MQ4_K_Mollama13.5 GiB on diskpublished by LiquidAI
LiquidAI/LFM2.5-24B-A2B-MLX-5bit5BITmlx15.4 GiB on diskpublished by LiquidAI
LiquidAI/LFM2.5-24B-A2B-FP8FP8vllm23.0 GiB on diskpublished by LiquidAI
LiquidAI/LFM2.5-24B-A2B-MLX-8bit8BITmlx23.7 GiB on diskpublished by LiquidAI
hf.co/LiquidAI/LFM2.5-24B-A2B-GGUF:Q8_0Q8_0ollama23.7 GiB on diskpublished by LiquidAI
LiquidAI/LFM2.5-24B-A2BBF16vllm44.7 GiB on diskpublished by LiquidAI
LiquidAI/LFM2.5-24B-A2B-MLX-bf16BF16mlx44.7 GiB on diskpublished by LiquidAI
hf.co/LiquidAI/LFM2.5-24B-A2B-GGUF:BF16BF16ollama44.7 GiB on diskpublished by LiquidAI
hf.co/LiquidAI/LFM2.5-24B-A2B-GGUF:F16F16ollama44.7 GiB on diskpublished by LiquidAI
LFM2.5 24B-A2B-BASE
LiquidAI/LFM2.5-24B-A2B-BaseBF16vllm44.7 GiB on diskpublished by LiquidAI
LFM2.5-24B-A2B-Uncensored-GGUF
hf.co/AtomicChat/LFM2.5-24B-A2B-Uncensored-GGUF:IQ3_MIQ3_Mollama11.3 GiB on diskderived by AtomicChat
hf.co/AtomicChat/LFM2.5-24B-A2B-Uncensored-GGUF:IQ4_XSIQ4_XSollama13.5 GiB on diskderived by AtomicChat
hf.co/AtomicChat/LFM2.5-24B-A2B-Uncensored-GGUF:Q4_K_MQ4_K_Mollama14.3 GiB on diskderived by AtomicChat
hf.co/AtomicChat/LFM2.5-24B-A2B-Uncensored-GGUF:Q5_K_MQ5_K_Mollama16.1 GiB on diskderived by AtomicChat
hf.co/AtomicChat/LFM2.5-24B-A2B-Uncensored-GGUF:Q8_0Q8_0ollama24.0 GiB on diskderived by AtomicChat
LFM2.5-24B-A2B
unsloth/LFM2.5-24B-A2BBF16vllm44.7 GiB on diskderived by unsloth
LFM2.5-8B-A1B-Qwen3.8-Turbo-Brilliance-Power-X12-NEO-MAX-GGUF
hf.co/DavidAU/LFM2.5-8B-A1B-Qwen3.8-Turbo-Brilliance-Power-X12-NEO-MAX-GGUF:IQ4_XSIQ4_XSollama4.6 GiB on diskderived by DavidAU
hf.co/DavidAU/LFM2.5-8B-A1B-Qwen3.8-Turbo-Brilliance-Power-X12-NEO-MAX-GGUF:Q6_KQ6_Kollama6.8 GiB on diskderived by DavidAU
hf.co/DavidAU/LFM2.5-8B-A1B-Qwen3.8-Turbo-Brilliance-Power-X12-NEO-MAX-GGUF:Q8_0Q8_0ollama8.6 GiB on diskderived by DavidAU
hf.co/DavidAU/LFM2.5-8B-A1B-Qwen3.8-Turbo-Brilliance-Power-X12-NEO-MAX-GGUF:BF16BF16ollama15.8 GiB on diskderived by DavidAU
LFM2.5 DSPARK-24B-A2B
LiquidAI/LFM2.5-24B-A2B-DSparkBF16vllm625 MiB on diskpublished by LiquidAI
LFM2.5 VL-3B
hf.co/unsloth/LFM2.5-VL-3B-GGUF:IQ2_MIQ2_Mollama961 MiB on diskquantized by unsloth
hf.co/unsloth/LFM2.5-VL-3B-GGUF:IQ3_XXSIQ3_XXSollama1.1 GiB on diskquantized by unsloth
hf.co/unsloth/LFM2.5-VL-3B-GGUF:IQ4_XSIQ4_XSollama1.4 GiB on diskquantized by unsloth
hf.co/unsloth/LFM2.5-VL-3B-GGUF:Q4_0Q4_0ollama1.5 GiB on diskquantized by unsloth
hf.co/unsloth/LFM2.5-VL-3B-GGUF:Q4_K_MQ4_K_Mollama1.6 GiB on diskquantized by unsloth
hf.co/unsloth/LFM2.5-VL-3B-GGUF:Q8_0Q8_0ollama2.7 GiB on diskquantized by unsloth
hf.co/unsloth/LFM2.5-VL-3B-GGUF:BF16BF16ollama5.0 GiB on diskquantized by unsloth
LiquidAI/LFM2.5-VL-3B-MLX-bf16BF16mlx5.8 GiB on diskpublished by LiquidAI
LiquidAI/LFM2.5-VL-3BBF16vllm5.8 GiB on diskpublished by LiquidAI
LFM2.5 VL-DSPARK-3B
LiquidAI/LFM2.5-VL-3B-DSparkBF16vllm533 MiB on diskpublished by LiquidAI
ask your own gram

Your concierge can answer this from the network rather than from this page. The ask carries the operator id, so it resolves to a capability and an engine — not to a name somebody has to recognise.

What has this network measured about the LFM2.5 family (epn.family.lfm-2-5), and can my gram run any of it? See /inferences/lfm-2-5

open a gramx with this ask →epn.family.lfm-2-5