GLM-4.7-Flash has 31.2B parameters (3.6B active per token). In BF16 its weights alone take 55.8 GiB; quantized to 4-bit, about 16.9 GiB. Each 32K-token request adds 1.65 GiB of KV cache. Cheapest way to run it today: 1 × GeForce RTX 3090 (24 GB) at $0.108 per hour. It calls tools and has a thinking mode you can switch off.
Read every day from Hugging Face. Split files are added together; vision projectors are left out.
How many GPUs to run GLM-4.7-Flash
Fewest GPUs that fit the weights plus one 8K-token request. Data-center GPUs use vLLM (90% of memory usable), the others llama.cpp. Several GPUs split the model with tensor or pipeline parallelism.
GPU
BF16 / FP16
8-bit
4-bit
GeForce RTX 4090 24 GB
3
2
1
GeForce RTX 5090 32 GB
2
1
1
RTX PRO 6000 Blackwell 96 GB
1
1
1
Mac M5 Max 128 GB 96 GB usable
1
1
1
Ryzen AI Max+ 395 128 GB 96 GB usable
1
1
1
DGX Spark (GB10) 128 GB
1
1
1
NVIDIA H100 80 GB
1
1
1
NVIDIA H200 141 GB
1
1
1
NVIDIA B200 (HGX) 180 GB
1
1
1
8-bit: FP8 with vLLM (INT8 on GPUs without FP8), Q8_0 with llama.cpp. 4-bit: AWQ / GPTQ with vLLM, Q4_K_M with llama.cpp, both counted at the Q4_K_M size.
Cheapest way to run GLM-4.7-Flash today
For each GPU: the fewest cards that fit GLM-4.7-Flash with one 8K-token request, the most faithful weight format at that count, and the cheapest on-demand price (2026-11-06).
On-demand prices from AWS, Google Cloud, Azure, RunPod, Vast.ai, Oracle Cloud, OVHcloud, Scaleway, Verda, Akash and UpCloud, checked every hour. Serving many users needs more KV cache, so more memory: size it in the calculator. Some links are affiliate links: StudioTV may earn a commission, at no extra cost to you.
vLLM and SGLang apply these defaults unless a request sets its own. With llama.cpp, pass them yourself: --temp 1
Reasoning. Thinking mode you can switch on or off. With vLLM or SGLang, set it per request: "chat_template_kwargs": {"enable_thinking": false} turns it off, true on. Reasoning tokens take context and KV cache like any other token: leave room for them.
KV cache: how memory grows with context
Architecture: 47 layers · MLA (512 + 64 latent) on 47. Its cache stays compact because MLA stores one compressed latent per token and layer instead of full keys and values. Only 3.6B parameters are active per token, but all 31B must sit in memory: VRAM depends on the total, speed on the active part.
Context per request
BF16 cache
FP8 cache
8K tokens
0.41 GiB
0.21 GiB
32K tokens
1.65 GiB
0.83 GiB
128K tokens
6.61 GiB
3.30 GiB
198K tokens
10.2 GiB
5.11 GiB
GLM-4.7-Flash VRAM FAQ
How much VRAM does GLM-4.7-Flash need?
In BF16 the weights alone take 55.8 GiB (60 GB). In FP8 that is 28.5 GiB, and about 16.9 GiB with 4-bit quantization (Q4_K_M). Each request then adds KV cache: 0.41 GiB at 8K tokens and 1.65 GiB at 32K (BF16 cache).
Can GLM-4.7-Flash run on a single RTX 4090 (24 GB)?
Yes. In Q4_K_M it fits on one RTX 4090 with an 8K-token context (17.3 GiB including the cache, with llama.cpp).
Can GLM-4.7-Flash run on an RTX 5090 (32 GB)?
Yes. In Q8_0 (28.9 GiB with an 8K-token context) it fits on one RTX 5090, with llama.cpp.
Can GLM-4.7-Flash run on a Mac?
Yes. In 4-bit (Q4_K_M with llama.cpp) it needs a Mac with 32 GB of unified memory or more, of which macOS gives the GPU about 21 GB, for one request with an 8K-token context. In 8-bit (Q8_0) it needs 48 GB or more.
Can GLM-4.7-Flash run on a DGX Spark or a Ryzen AI Max+ 395 (128 GB)?
Yes. In BF16 (56.2 GiB with an 8K-token context) it fits on a DGX Spark (128 GB) and on a Ryzen AI Max+ 395 with 128 GB, whose GPU gets up to 96 GB, with llama.cpp and one request.
What is the cheapest way to run GLM-4.7-Flash?
On 2026-11-06, the cheapest on-demand setup is 1 × GeForce RTX 3090 (24 GB) with Q4_K_M weights on Vast.ai, at $0.108 per hour (about $78.84 per month). Next: 2 × RTX A4000 (16 GB) with Q4_K_M weights on Vast.ai, at $0.144 per hour. Sized for one 8K-token request; prices are checked every hour.
How many H100 GPUs does GLM-4.7-Flash need?
With one request and an 8K-token context: 1 in BF16, 1 in FP8 and 1 in 4-bit (vLLM, 90% of memory usable). Serving many users at once needs more memory for their KV caches: the calculator sizes that for you.
Does GLM-4.7-Flash support tool calling?
Yes, GLM-4.7-Flash supports tool (function) calling: it can ask your application to run a function and use the result.
Does GLM-4.7-Flash have a thinking mode?
Thinking mode you can switch on or off. With vLLM or SGLang, set it per request with "chat_template_kwargs": {"enable_thinking": false} (or true). Reasoning tokens count toward the context and the KV cache.
What temperature should I use with GLM-4.7-Flash?
The official defaults are temperature 1.
GLM-4.7-Flash VRAM badge
For a model card or a README: the 4-bit size and the smallest GPU it fits on, linked to this page. Data as JSON: /api/models/glm-4-7-flash.json.
Estimates, not guarantees: computed from the official config.json with the same engine as the LLM VRAM Calculator. Real usage depends on your engine version and settings.
Z.ai, China · Released January 19, 2026