GL

Glm-4.1V-9B-Thinking

GLMGGUF by unslothUpdated 2025-07-25
Download on Hugging Face
Parameters9.4B
Context64K
Layers40
Architectureglm4
Smallest quant3.2 GB
Quick answer

Glm-4.1V-9B-Thinking needs about 7.0 GB of memory at Q4_K_M (a 6.2 GB file) with an 8K context window, so it fits on a 8 GB GPU. This repository has 26 quants, from 3.2 GB (IQ1_S) to 18.8 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 1.3 GB.

GLM‑4.1V‑9B‑Thinking – a 9B vision‑language model with advanced reasoning

GLM‑4.1V‑9B‑Thinking is a 9‑billion‑parameter VLM that supports 64k token context and up to 4K image resolution. It is designed for reasoning tasks and offers bilingual Chinese‑English support. The model is released under the MIT license.

Good at

  • Reasoning across math and complex tasks
  • Vision‑language understanding with arbitrary aspect ratios
  • Long‑context comprehension up to 64k tokens

Limits

  • Requires large VRAM for full precision (≈10 GB for Q8_0)
  • No native vision support in llama.cpp without custom template
  • Limited to 8‑k context for memory estimates in the table

Best for

  • Complex problem solving with images
  • Multimodal agent development
  • Long‑form reasoning in bilingual settings

Who it’s for. Researchers and developers needing a high‑capacity VLM for reasoning tasks. The model is suitable for academic projects and prototyping of multimodal agents.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S3.2 GB 4.0 GB
Fits Get file →
IQ1_M3.3 GB 4.2 GB
Fits Get file →
IQ2_XXS3.5 GB 4.3 GB
Fits Get file →
IQ2_M4.0 GB 4.8 GB
Fits Get file →
Q2_K4.0 GB 4.8 GB
Fits Get file →
Q2_K_L4.2 GB 5.0 GB
Fits Get file →
Q2_K_XL4.2 GB 5.0 GB
Fits Get file →
IQ3_XXS4.2 GB 5.1 GB
Fits Get file →
Q3_K_S4.6 GB 5.4 GB
Fits Get file →
Q3_K_M5.0 GB 5.8 GB
Fits Get file →
Q3_K_XL5.1 GB 6.0 GB
Fits Get file →
IQ4_XS5.3 GB 6.1 GB
Fits Get file →
IQ4_NL5.5 GB 6.3 GB
Fits Get file →
Q4_05.5 GB 6.3 GB
Fits Get file →
Q4_K_S5.8 GB 6.6 GB
Fits Get file →
Q4_16.0 GB 6.8 GB
Fits Get file →
Q4_K_M6.2 GB 7.0 GB
Fits Get file →
Q4_K_XL6.2 GB 7.0 GB
Fits Get file →
Q5_K_S6.7 GB 7.5 GB
Tight Get file →
Q5_K_XL7.0 GB 7.8 GB
Tight Get file →
Q5_K_M7.1 GB 7.9 GB
Tight Get file →
Q6_K8.3 GB 9.1 GB
Too big Get file →
Q6_K_XL8.9 GB 9.7 GB
Too big Get file →
Q8_010.0 GB 10.8 GB
Too big Get file →
Q8_K_XL12.1 GB 12.9 GB
Too big Get file →
BF1618.8 GB 19.6 GB
Too big Get file →

For 8 GB VRAM, use Q8_0 (10 GB file, ~10.8 GB memory at 8k context). If 12 GB VRAM is available, Q8_K_XL (12.08 GB file, ~12.9 GB memory) offers higher precision. For 4 GB VRAM, Q4_0 (5.48 GB file, ~6.3 GB memory) is the most efficient option.

What GPU do you need for Glm-4.1V-9B-Thinking?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: the best quant that fits is IQ1_S (3.2 GB file, about 4.0 GB in use).
  • 6 GB VRAM: the best quant that fits is Q3_K_XL (5.1 GB file, about 6.0 GB in use).
  • 8 GB VRAM: the best quant that fits is Q5_K_M (7.1 GB file, about 7.9 GB in use).
  • 12 GB VRAM: the best quant that fits is Q8_0 (10.0 GB file, about 10.8 GB in use).
  • 16 GB VRAM: the best quant that fits is Q8_K_XL (12.1 GB file, about 12.9 GB in use).
  • 24 GB VRAM: the best quant that fits is BF16 (18.8 GB file, about 19.6 GB in use).

Which Glm-4.1V-9B-Thinking quant should you choose?

Q4_K_M is the usual starting point for Glm-4.1V-9B-Thinking: it balances file size and quality well. Here it is a 6.2 GB file.

The Q8_0 file is 10.0 GB, about 1.6 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is IQ1_S at 3.2 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureglm4
Parameters9.4B
Layers40
Trained context65,536 tokens
KV cache (fp16)40 KB per token

Source

Quantized byunsloth
FamilyGLM
Licensemit
Downloads (HF)4,294
Last updated2025-07-25

Frequently asked questions about Glm-4.1V-9B-Thinking

How much VRAM does Glm-4.1V-9B-Thinking need?

About 7.0 GB at Q4_K_M with an 8K context window. That is the 6.2 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run Glm-4.1V-9B-Thinking on an 8 GB GPU?

Yes, with Q5_K_M: the 7.1 GB file needs about 7.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Glm-4.1V-9B-Thinking GGUF?

It depends on the quant: IQ1_S is 3.2 GB, IQ1_M is 3.3 GB, IQ2_XXS is 3.5 GB, IQ2_M is 4.0 GB, Q2_K is 4.0 GB, Q2_K_L is 4.2 GB and 20 more.

What context length does Glm-4.1V-9B-Thinking support?

The GGUF header lists a trained context length of 65,536 tokens (64K). The KV cache grows with the context you use: about 0.3 GB at 8K and 1.3 GB at 32K.

Where can I download Glm-4.1V-9B-Thinking GGUF?

The files are on Hugging Face in the unsloth/GLM-4.1V-9B-Thinking-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Glm-4.1V-9B-Thinking use?

Hugging Face lists the license as mit. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run GLM‑4.1V‑9B‑Thinking on 8 GB VRAM?

Yes, using the Q8_0 quantization which requires about 10.8 GB memory at 8k context. It will fit on 8 GB VRAM with some swapping or lower context.

Does the model support image input in llama.cpp?

Image support requires the custom chat template fixes from Unsloth. Without them, llama.cpp will not process images.

What context length can I use for inference?

The model supports up to 64k tokens, but memory estimates in the table are for 8k context. Larger contexts will increase memory usage proportionally.

Other GGUFs of GLM-4.1V-9B-Thinking

Similar models

Browse all →

More from unsloth

All by unsloth →

More GLM models

All GLM →

Other small models

All small models →