model family
GLM-5 logo

GLM-5

Published by zai-org. 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 20 builds measured · 19 still open
GLM-5_4layer1 build measured →
hf.co/mradermacher/GLM-5_4layer-i1-GGUF:Q4_K_M4layer~pinaster · Q4_K_M · mradermacher · ollama · 1 machine
—recalled
—served
7.7best tok/s

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

6 of 20 builds fit a 16 GB machine — budgeting 70% of it, or 11.2 GiB, because the cache and the runtime want the rest.

  • tiny-random/glm-5BF16vllmderived by tiny-random
    0.0 GiB
  • hf.co/mradermacher/GLM-5_4layer-i1-GGUF:Q2_KQ2_Kollamaderived by mradermacher
    4.8 GiB
  • hf.co/mradermacher/GLM-5_4layer-i1-GGUF:Q3_K_SQ3_K_Sollamaderived by mradermacher
    5.6 GiB
  • hf.co/mradermacher/GLM-5_4layer-i1-GGUF:IQ3_SIQ3_Sollamaderived by mradermacher
    5.6 GiB
  • hf.co/mradermacher/GLM-5_4layer-i1-GGUF:Q4_0Q4_0ollamaderived by mradermacher
    7.0 GiB
  • hf.co/mradermacher/GLM-5_4layer-i1-GGUF:Q4_K_MQ4_K_Mollamameasured herederived by mradermacher
    7.8 GiB
  • hf.co/mradermacher/GLM-5_4layer-GGUF:Q8_0Q8_0ollamaderived by mradermacher
    12.9 GiB
  • unsloth/GLM-5-GGUF:IQ1_SIQ1_Sllamacppquantized by unsloth
    189.7 GiB
  • unsloth/GLM-5-GGUF:IQ2_XXSIQ2_XXSllamacppquantized by unsloth
    224.5 GiB
  • unsloth/GLM-5-GGUF:IQ3_XXSIQ3_XXSllamacppquantized by unsloth
    283.6 GiB
  • mlx-community/GLM-5-4bit4BITmlxquantized by mlx-community
    389.9 GiB
  • amd/GLM-5-MXFP4QUARKvllmquantized by amd
    395.5 GiB
  • unsloth/GLM-5-GGUF:Q4_0Q4_0llamacppquantized by unsloth
    397.8 GiB
  • unsloth/GLM-5-GGUF:Q4_K_MQ4_K_Mllamacppquantized by unsloth
    424.6 GiB
  • nvidia/GLM-5-NVFP4BF16vllmderived by nvidia
    447.7 GiB
  • zai-org/GLM-5-FP8FP8vllmpublished by zai-org
    704.2 GiB
  • mlx-community/GLM-5-8bit-MXFP88BITmlxderived by mlx-community
    714.6 GiB
  • unsloth/GLM-5-GGUF:Q8_0Q8_0llamacppquantized by unsloth
    746.3 GiB
  • zai-org/GLM-5BF16vllmpublished by zai-org
    1404.2 GiB
  • unsloth/GLM-5-GGUF:BF16BF16llamacppquantized by unsloth
    1404.4 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
GLM-5_4layer
hf.co/mradermacher/GLM-5_4layer-i1-GGUF:Q2_KQ2_Kollama4.8 GiB on diskderived by mradermacher
hf.co/mradermacher/GLM-5_4layer-i1-GGUF:Q3_K_SQ3_K_Sollama5.6 GiB on diskderived by mradermacher
hf.co/mradermacher/GLM-5_4layer-i1-GGUF:IQ3_SIQ3_Sollama5.6 GiB on diskderived by mradermacher
hf.co/mradermacher/GLM-5_4layer-i1-GGUF:Q4_0Q4_0ollama7.0 GiB on diskderived by mradermacher
hf.co/mradermacher/GLM-5_4layer-i1-GGUF:Q4_K_MQ4_K_Mollama7.8 GiB on diskderived by mradermacher
hf.co/mradermacher/GLM-5_4layer-GGUF:Q8_0Q8_0ollama12.9 GiB on diskderived by mradermacher
GLM-5 BASE
unsloth/GLM-5-GGUF:IQ1_SIQ1_Sllamacpp189.7 GiB on diskquantized by unsloth
unsloth/GLM-5-GGUF:IQ2_XXSIQ2_XXSllamacpp224.5 GiB on diskquantized by unsloth
unsloth/GLM-5-GGUF:IQ3_XXSIQ3_XXSllamacpp283.6 GiB on diskquantized by unsloth
mlx-community/GLM-5-4bit4BITmlx389.9 GiB on diskquantized by mlx-community
amd/GLM-5-MXFP4QUARKvllm395.5 GiB on diskquantized by amd
unsloth/GLM-5-GGUF:Q4_0Q4_0llamacpp397.8 GiB on diskquantized by unsloth
unsloth/GLM-5-GGUF:Q4_K_MQ4_K_Mllamacpp424.6 GiB on diskquantized by unsloth
zai-org/GLM-5-FP8FP8vllm704.2 GiB on diskpublished by zai-org
unsloth/GLM-5-GGUF:Q8_0Q8_0llamacpp746.3 GiB on diskquantized by unsloth
zai-org/GLM-5BF16vllm1404.2 GiB on diskpublished by zai-org
unsloth/GLM-5-GGUF:BF16BF16llamacpp1404.4 GiB on diskquantized by unsloth
GLM-5-8bit-MXFP8
mlx-community/GLM-5-8bit-MXFP88BITmlx714.6 GiB on diskderived by mlx-community
GLM-5-NVFP4
nvidia/GLM-5-NVFP4BF16vllm447.7 GiB on diskderived by nvidia
glm-5
tiny-random/glm-5BF16vllm8 MiB on diskderived by tiny-random
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 GLM-5 family (epn.family.glm-5), and can my gram run any of it? See /inferences/glm-5

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