readygpu

GLM 4.6V Flash

10.3B parameters · 128k context · MIT · released 2025-12-07

visionagenticcodingreasoningchat

In short

GLM 4.6V Flash is an open model from Z.ai with 10.3B parameters. At Q4_K_M, GLM 4.6V Flash needs about 7 GB of memory at 8k context and 8 GB at 32k, counting the weights, the KV cache and runtime buffers. The cheapest hardware in the UK that runs GLM 4.6V Flash fully in memory at 32k context is the NVIDIA Tesla P40, £261 at eBay, at about 19 tok/s (estimated).

Updated . Speeds are estimates; how they are worked out.

What to run it on in the UK

Fully in memory at 32k context, priced at retailers today.

All hardware that runs it

Everything we price, best value first. Two cards split the model by layers.

HardwareQuantFitsDecodePricetok/s per 1,000
NVIDIA Tesla P40Q8_0fits19 tok/s£26172.5
Intel Arc B580Q4_K_Mfits30 tok/s£32394.1
AMD Radeon RX 7600 XTQ8_0fits14 tok/s£34840.2
NVIDIA GeForce RTX 3060 12GBQ4_K_Mfits30 tok/s£39076.2
NVIDIA GeForce RTX 3080 10GBQ4_K_Mfits63 tok/s£466134.9
AMD Radeon RX 9060 XT 16GBQ8_0fits16 tok/s£49931.1
NVIDIA GeForce RTX 5060 Ti 16GBQ8_0fits25 tok/s£55044.5
AMD Radeon RX 7800 XTQ8_0fits30 tok/s£56853.4
Intel Arc A770 16GBQ8_0fits25 tok/s£58842.0
AMD Radeon RX 9070 XTQ8_0fits31 tok/s£67046.3
NVIDIA GeForce RTX 5070Q4_K_Mfits56 tok/s£69579.9
Intel Arc Pro B60Q8_0fits20 tok/s£74227.1
Mac mini M4 16GBQ4_K_Mfits10 tok/s£76513.6
AMD Radeon RX 7900 XTQ8_0fits39 tok/s£78949.2
NVIDIA GeForce RTX 4070 SuperQ4_K_Mfits42 tok/s£87447.6
NVIDIA GeForce RTX 4070 Ti SuperQ8_0fits37 tok/s£95038.6
NVIDIA GeForce RTX 5070 TiQ8_0fits49 tok/s£97550.3
NVIDIA GeForce RTX 4060 Ti 16GBQ8_0fits16 tok/s£97716.1
AMD Radeon RX 7900 XTXQ8_0fits47 tok/s£99946.6
NVIDIA GeForce RTX 5080Q8_0fits53 tok/s£1,19943.8
Mac mini M4 32GBQ8_0fits7 tok/s£1,2005.8
NVIDIA GeForce RTX 4080 SuperQ8_0fits40 tok/s£1,42928.1
AMD Radeon AI PRO R9700Q8_0fits31 tok/s£1,67418.5
NVIDIA GeForce RTX 3090Q8_0fits51 tok/s£1,70730.0
NVIDIA GeForce RTX 3090 TiQ8_0fits55 tok/s£2,19925.1
Mac mini M4 Pro 48GBQ8_0fits16 tok/s£3,0005.2
GMKtec EVO-X2 Ryzen AI Max+ 395 128GBQ8_0fits12 tok/s£3,3203.7
Mac Studio M4 Max 64GBQ8_0fits31 tok/s£3,3999.2
NVIDIA GeForce RTX 5090Q8_0fits98 tok/s£5,00019.6
Mac Studio M3 Ultra 96GBQ8_0fits47 tok/s£5,4998.5
NVIDIA GeForce RTX 4090Q8_0fits55 tok/s£6,0159.2
Mac Studio M3 Ultra 256GBQ8_0fits47 tok/s£8,4995.5
Mac Studio M3 Ultra 512GBQ8_0fits47 tok/s£17,4952.7
NVIDIA RTX PRO 6000 BlackwellQ8_0fits98 tok/s£29,5043.3
Mac Studio M4 Max 128GBQ8_0fits31 tok/s––
Mac mini M4 Pro 64GBQ8_0fits16 tok/s––
Framework Desktop Ryzen AI Max+ 395 128GBQ8_0fits12 tok/s––
NVIDIA DGX Spark 128GBQ8_0fits15 tok/s––
ASUS Ascent GX10 128GBQ8_0fits15 tok/s––
Framework Desktop Ryzen AI Max+ 395 64GBQ8_0fits12 tok/s––

Memory needed

Weights, KV cache and runtime buffers at each context length (KV cache at fp16).

QuantFile@8k@32k@64k@128k
Q4_K_Mgguf5.7 GB6.8 GB7.9 GB9.3 GB12.2 GB
Q8_0gguf9.3 GB10.4 GB11.5 GB12.9 GB15.8 GB

Run it

Copy a command for the runtime you use.

  • Q4_K_M
    • Ollamaollama run hf.co/ggml-org/GLM-4.6V-Flash-GGUF:Q4_K_M
    • LM Studiolms get ggml-org/GLM-4.6V-Flash-GGUF@q4_k_m
    • llama.cppllama-server -hf ggml-org/GLM-4.6V-Flash-GGUF:Q4_K_M
  • Q8_0
    • Ollamaollama run hf.co/ggml-org/GLM-4.6V-Flash-GGUF:Q8_0
    • LM Studiolms get ggml-org/GLM-4.6V-Flash-GGUF@q8_0
    • llama.cppllama-server -hf ggml-org/GLM-4.6V-Flash-GGUF:Q8_0

Notes

This repo hosts GLM-4.6V-Flash (9B), the lightweight variant, not the 106B GLM-4.6V; vLLM serving needs vllm>=0.12.0 and transformers>=5.0.0rc0 (SGLang needs >=0.5.6.post1). The card admits that pure-text QA is weak and that the model can overthink or repeat itself.

Questions

How much VRAM does GLM 4.6V Flash need?
At Q4_K_M, GLM 4.6V Flash needs about 7 GB of memory at 8k context and 8 GB at 32k, counting the weights, the KV cache and runtime buffers.
What is the best GPU for running GLM 4.6V Flash locally?
The cheapest hardware in the UK that runs GLM 4.6V Flash fully in memory at 32k context is the NVIDIA Tesla P40, £261 at eBay, at about 19 tok/s (estimated). The best value is the NVIDIA GeForce RTX 3080 10GB (about 63 tok/s for £466). The fastest we price is the NVIDIA GeForce RTX 5090, at about 98 tok/s.
Can I run GLM 4.6V Flash on a 16 GB GPU?
Yes: at Q8_0 it fits fully in 16 GB at 32k context. On the NVIDIA GeForce RTX 5080 it runs at about 53 tok/s (estimated).
Can I run GLM 4.6V Flash on a 24 GB GPU?
Yes: at Q8_0 it fits fully in 24 GB at 32k context. On the NVIDIA GeForce RTX 3090 Ti it runs at about 55 tok/s (estimated).
How do I run GLM 4.6V Flash locally?
With Ollama: ollama run hf.co/ggml-org/GLM-4.6V-Flash-GGUF:Q4_K_M (the Q4_K_M build). LM Studio and llama.cpp commands are on this page.
What licence is GLM 4.6V Flash released under?
MIT, which allows commercial use.