GLM · 9B to 32B · Mixture of experts
GLM 4.7 Flash
31.22 billion parameters in total, but only 3.66 billion are read for each token. That split is the whole point of the design: it costs the memory of a large model and the speed of a small one. It compresses its cache into a latent vector, which keeps the cache small at long context.
About GLM 4.7 Flash
GLM 4.7 Flash is a mixture-of-experts model with 31.22B parameters, 3.66B of them active for each token, built from 47 layers with a hidden size of 2048. Its repository was published on 2026-01-19. The license listed on Hugging Face is mit. It compresses its cache into a latent vector, which keeps the cache small at long context.
Memory needed, by quantization
At an 8k context window. Quality is our rough ranking of how much the compression costs you: anything at or above Q5 is hard to tell apart from the original in normal use.
| Quantization | File size | KV cache | Total VRAM | Quality | What it costs you |
|---|---|---|---|---|---|
| Q8_0 | 30.89 GB | 0.41 GB | 31.33 GB | 99% | Lossless in practice. Use it when the memory is there. |
| Q6_K | 23.84 GB | 0.41 GB | 24.35 GB | 98% | Very close to Q8 for two thirds of the size. |
| Q5_K_M | 20.68 GB | 0.41 GB | 21.22 GB | 96% | The quality-first choice when Q6 will not fit. |
| Q4_K_M | 17.55 GB | 0.41 GB | 18.13 GB | 93% | The default. Best size-to-quality ratio for local use. |
| IQ4_XS | 15.45 GB | 0.41 GB | 16.04 GB | 91% | Importance-matrix 4-bit. Q4_K_S quality, smaller file. |
| Q3_K_M | 14.21 GB | 0.41 GB | 14.82 GB | 86% | Degradation starts to show. A way to fit one size up. |
| IQ3_XXS | 11.12 GB | 0.41 GB | 11.76 GB | 79% | Aggressive. Only worth it on very large models. |
What a longer conversation costs
Same model at Q4_K_M, only the context window changes. This model compresses its cache into a latent vector, which is why the numbers stay small at long context.
| Context | KV cache | Total VRAM | Fits in 8 GB | Fits in 12 GB | Fits in 24 GB |
|---|---|---|---|---|---|
| 4k | 0.21 GB | 17.92 GB | no | no | yes |
| 8k | 0.41 GB | 18.13 GB | no | no | yes |
| 16k | 0.83 GB | 18.54 GB | no | no | yes |
| 32k | 1.65 GB | 19.37 GB | no | no | yes |
| 64k | 3.3 GB | 21.02 GB | no | no | yes |
| 128k | 6.61 GB | 24.32 GB | no | no | no |
The "fits" columns allow for the roughly 0.8 GB Windows keeps for the desktop.
Which hardware runs GLM 4.7 Flash
55 of 118 consumer devices run it at a quantization worth using, at an 8k context window.
| Device | Memory | Verdict | Quantization | Used | Speed |
|---|---|---|---|---|---|
| RTX PRO 6000 Blackwell NVIDIA | 96 GB | Runs great | Q8_0 | 31.33 GB | 191.3 t/s |
| RTX 6000 Ada Generation NVIDIA | 48 GB | Runs great | Q8_0 | 31.33 GB | 102.5 t/s |
| RTX A6000 NVIDIA | 48 GB | Runs great | Q8_0 | 31.33 GB | 82 t/s |
| Apple M3 Ultra 96GB Apple | 96 GB | Runs great | Q8_0 | 31.61 GB | 80.8 t/s |
| Apple M3 Ultra 256GB Apple | 256 GB | Runs great | Q8_0 | 31.61 GB | 80.8 t/s |
| Apple M3 Ultra 512GB Apple | 512 GB | Runs great | Q8_0 | 31.61 GB | 80.8 t/s |
| Apple M1 Ultra 64GB Apple | 64 GB | Runs great | Q8_0 | 31.61 GB | 78.9 t/s |
| Apple M1 Ultra 128GB Apple | 128 GB | Runs great | Q8_0 | 31.61 GB | 78.9 t/s |
| Apple M2 Ultra 64GB Apple | 64 GB | Runs great | Q8_0 | 31.61 GB | 78.9 t/s |
| Apple M2 Ultra 128GB Apple | 128 GB | Runs great | Q8_0 | 31.61 GB | 78.9 t/s |
| Apple M2 Ultra 192GB Apple | 192 GB | Runs great | Q8_0 | 31.61 GB | 78.9 t/s |
| Radeon PRO W7900 AMD | 48 GB | Runs great | Q8_0 | 31.33 GB | 78.2 t/s |
| Apple M4 Max 48GB Apple | 48 GB | Runs great | Q8_0 | 31.61 GB | 53.9 t/s |
| Apple M4 Max 64GB Apple | 64 GB | Runs great | Q8_0 | 31.61 GB | 53.9 t/s |
| Apple M4 Max 128GB Apple | 128 GB | Runs great | Q8_0 | 31.61 GB | 53.9 t/s |
| Apple M1 Max 64GB Apple | 64 GB | Runs great | Q8_0 | 31.61 GB | 39.5 t/s |
| Apple M2 Max 64GB Apple | 64 GB | Runs great | Q8_0 | 31.61 GB | 39.5 t/s |
| Apple M2 Max 96GB Apple | 96 GB | Runs great | Q8_0 | 31.61 GB | 39.5 t/s |
| Apple M3 Max 48GB Apple | 48 GB | Runs great | Q8_0 | 31.61 GB | 39.5 t/s |
| Apple M3 Max 64GB Apple | 64 GB | Runs great | Q8_0 | 31.61 GB | 39.5 t/s |
| Apple M3 Max 96GB Apple | 96 GB | Runs great | Q8_0 | 31.61 GB | 39.5 t/s |
| Apple M3 Max 128GB Apple | 128 GB | Runs great | Q8_0 | 31.61 GB | 39.5 t/s |
| NVIDIA DGX Spark 128GB NVIDIA | 128 GB | Runs great | Q8_0 | 31.64 GB | 29.1 t/s |
| Apple M4 Pro 48GB Apple | 48 GB | Runs great | Q8_0 | 31.61 GB | 26.9 t/s |
| Apple M4 Pro 64GB Apple | 64 GB | Runs great | Q8_0 | 31.61 GB | 26.9 t/s |
Show 35 more devices
| Jetson AGX Orin 64GB NVIDIA | 64 GB | Runs well | Q8_0 | 31.64 GB | 21.9 t/s |
| Ryzen AI Max+ 395 64GB AMD | 64 GB | Runs well | Q8_0 | 31.64 GB | 21.1 t/s |
| Ryzen AI Max+ 395 96GB AMD | 96 GB | Runs well | Q8_0 | 31.64 GB | 21.1 t/s |
| Ryzen AI Max+ 395 128GB AMD | 128 GB | Runs well | Q8_0 | 31.64 GB | 21.1 t/s |
| GeForce RTX 5090 NVIDIA | 32 GB | Runs great | Q6_K | 24.35 GB | 240.5 t/s |
| GeForce RTX 4090 NVIDIA | 24 GB | Runs great | Q5_K_M | 21.22 GB | 152.9 t/s |
| GeForce RTX 3090 Ti NVIDIA | 24 GB | Runs great | Q5_K_M | 21.22 GB | 152.9 t/s |
| GeForce RTX 3090 NVIDIA | 24 GB | Runs great | Q5_K_M | 21.22 GB | 142 t/s |
| GeForce RTX 5090 Laptop NVIDIA | 24 GB | Runs great | Q5_K_M | 21.22 GB | 135.9 t/s |
| Radeon RX 7900 XTX AMD | 24 GB | Runs great | Q5_K_M | 21.22 GB | 123.5 t/s |
| RTX A5000 NVIDIA | 24 GB | Runs great | Q5_K_M | 21.22 GB | 116.5 t/s |
| Radeon RX 7900 XT AMD | 20 GB | Runs great | Q4_K_M | 18.13 GB | 118.2 t/s |
| GeForce RTX 5080 NVIDIA | 16 GB | Just fits | Q3_K_M | 14.82 GB | 198.7 t/s |
| GeForce RTX 5070 Ti NVIDIA | 16 GB | Just fits | Q3_K_M | 14.82 GB | 185.4 t/s |
| GeForce RTX 5080 Laptop NVIDIA | 16 GB | Just fits | Q3_K_M | 14.82 GB | 158.9 t/s |
| GeForce RTX 4080 SUPER NVIDIA | 16 GB | Just fits | Q3_K_M | 14.82 GB | 152.3 t/s |
| GeForce RTX 4080 NVIDIA | 16 GB | Just fits | Q3_K_M | 14.82 GB | 148.4 t/s |
| GeForce RTX 4070 Ti SUPER NVIDIA | 16 GB | Just fits | Q3_K_M | 14.82 GB | 139.1 t/s |
| GeForce RTX 4090 Laptop NVIDIA | 16 GB | Just fits | Q3_K_M | 14.82 GB | 119.2 t/s |
| Radeon RX 9070 XT AMD | 16 GB | Just fits | Q3_K_M | 14.82 GB | 113.2 t/s |
| Radeon RX 9070 AMD | 16 GB | Just fits | Q3_K_M | 14.82 GB | 113.2 t/s |
| Radeon RX 7800 XT AMD | 16 GB | Just fits | Q3_K_M | 14.82 GB | 109.5 t/s |
| Radeon RX 7900 GRE AMD | 16 GB | Just fits | Q3_K_M | 14.82 GB | 101.1 t/s |
| GeForce RTX 5060 Ti 16GB NVIDIA | 16 GB | Just fits | Q3_K_M | 14.82 GB | 92.7 t/s |
| RTX A4000 NVIDIA | 16 GB | Just fits | Q3_K_M | 14.82 GB | 92.7 t/s |
| Radeon RX 6900 XT AMD | 16 GB | Just fits | Q3_K_M | 14.82 GB | 89.9 t/s |
| Radeon RX 6800 AMD | 16 GB | Just fits | Q3_K_M | 14.82 GB | 89.9 t/s |
| Arc A770 16GB Intel | 16 GB | Just fits | Q3_K_M | 14.82 GB | 89.5 t/s |
| GeForce RTX 4060 Ti 16GB NVIDIA | 16 GB | Just fits | Q3_K_M | 14.82 GB | 59.6 t/s |
| Radeon RX 7600 XT AMD | 16 GB | Just fits | Q3_K_M | 14.82 GB | 50.5 t/s |
| Apple M4 Max 36GB Apple | 36 GB | Just fits | n/a | 31.61 GB | 53.9 t/s |
| Apple M1 Max 32GB Apple | 32 GB | Just fits | n/a | 24.56 GB | 49.6 t/s |
| Apple M2 Max 32GB Apple | 32 GB | Just fits | n/a | 24.56 GB | 49.6 t/s |
| Apple M1 Pro 16GB Apple | 16 GB | Just fits | n/a ! | 11.84 GB | 46.3 t/s |
| Apple M2 Pro 16GB Apple | 16 GB | Just fits | n/a ! | 11.84 GB | 46.3 t/s |
How it compares
Other models in the same family or of a similar size, on a 12 GB card and a 24 GB card, each at the best quantization that fits.
| Model | Weights at 4 bits | GeForce RTX 3060 12GB | GeForce RTX 4090 |
|---|---|---|---|
| GLM 4.7 Flash | 17.6 GB | 37.6 t/s at Q4_K_M | 152.9 t/s at Q5_K_M |
| GLM 4.5 Air | 62.1 GB | does not fit well | 10.1 t/s at Q4_K_M |
| GLM 4.5 | 201.5 GB | does not fit well | does not fit well |
| GLM 5.2 | 423.6 GB | does not fit well | does not fit well |
| Gemma 4 31B Instruct | 17.6 GB | does not fit well | 35.5 t/s at Q4_K_M |
| Gemma 4 26B A4B Instruct | 14.5 GB | 37.6 t/s at Q4_K_M | 117 t/s at Q6_K |
What to buy to run GLM 4.7 Flash
The cheapest hardware that runs it at a quantization worth using and a speed you would not resent, at an 8k context window.
Cheapest that works
GeForce GTX 1650
$100 used
Q4_K_M · uses 18.13 GB of its 4 GB · about 17.8 tokens/s
Best value
Radeon RX 7900 XT
$650
Q4_K_M · about 118.2 tokens/s · 800 GB/s
| Device | Price | Memory | Runs it at | Speed | |
|---|---|---|---|---|---|
| GeForce GTX 1650 NVIDIA | $100 used | 4 GB | Q4_K_M | 17.8 t/s | Check price |
| GeForce GTX 1660 SUPER NVIDIA | $110 used | 6 GB | Q4_K_M | 26.8 t/s | Check price |
| GeForce GTX 1080 Ti NVIDIA | $150 used | 11 GB | Q4_K_M | 39.6 t/s | Check price |
| GeForce RTX 2060 12GB NVIDIA | $160 used | 12 GB | Q4_K_M | 36.4 t/s | Check price |
| GeForce RTX 2080 Ti NVIDIA | $250 used | 11 GB | Q4_K_M | 42.7 t/s | Check price |
| Arc A770 16GB Intel | $280 used | 16 GB | Q4_K_M | 57 t/s | Check price |
| Radeon RX 6800 AMD | $330 used | 16 GB | Q4_K_M | 57.1 t/s | Check price |
| GeForce RTX 5060 Ti 16GB NVIDIA | $430 | 16 GB | Q4_K_M | 58.3 t/s | Check price |
Indicative prices reviewed 2026-09-01; used prices are marketplace typical. Options that cost more than a cheaper one with no more memory and no more speed are hidden. Price links are Amazon affiliate links. Change the standard, the context window or the budget in the buying tool.
These numbers also cover
Fine-tunes share their base model's architecture, so they need exactly the same memory. If you are looking for one of these, the figures above apply unchanged.
unsloth/GLM-4.7-Flash
Where it sits in the catalog
Source: zai-org/GLM-4.7-Flash on Hugging Face. Downloaded 1.9M times in the last month. Published 2026-01-19. Architecture figures are read from the repository's own configuration file, so they move when the model does.
Questions people ask about GLM 4.7 Flash
How much VRAM does GLM 4.7 Flash need?
At Q4_K_M with an 8k context window, GLM 4.7 Flash needs about 18.13 GB: 17.38 GB of weights, 0.41 GB of KV cache and 0.34 GB of overhead. At the highest-quality file it needs about 31.33 GB.
What is the cheapest GPU that runs GLM 4.7 Flash?
The smallest consumer device in our catalog that fits it is the GeForce RTX 5080 with 16 GB, running it at Q3_K_M at roughly 198.7 tokens per second.
How much does a longer context window add for GLM 4.7 Flash?
On the memory side, a 32k window adds 1.7 GB of KV cache on top of the weights, and the cache grows more slowly than a conventional model because of its attention design. The longest window the model supports natively is 198k tokens.