GLM 4.7 Flash — 31B MoE.
Run GLM 4.7 Flash locally with Ollama — needs 19.4 GB VRAM at Q4_K_M:
ollama run glm-4.7-flash| Quant | Bits | VRAM @ 16K | Quality |
|---|---|---|---|
| IQ3_XXS | 3.25 | 13.7 GB 13.1 + 0.6 KV | low |
| IQ3_XS | 3.5 | 14.7 GB 14.1 + 0.6 KV | low |
| Q3_K_S | 3.64 | 15.2 GB 14.6 + 0.6 KV | low |
| IQ3_M | 3.76 | 15.7 GB 15.1 + 0.6 KV | low |
| Q3_K_M | 4 | 16.6 GB 16.0 + 0.6 KV | low |
| Q3_K_L | 4.3 | 17.8 GB 17.2 + 0.6 KV | moderate |
| IQ4_XS | 4.46 | 18.4 GB 17.8 + 0.6 KV | moderate |
| Q4_K_S | 4.67 | 19.2 GB 18.6 + 0.6 KV | moderate |
| Q4_K_M | 4.89 | 20.1 GB 19.4 + 0.6 KV | good |
| Q5_K_S | 5.57 | 22.7 GB 22.1 + 0.6 KV | good |
| Q5_K_M | 5.7 | 23.2 GB 22.6 + 0.6 KV | good |
| Q6_K | 6.56 | 26.5 GB 25.9 + 0.6 KV | excellent |
| Q8_0 | 8.5 | 34.0 GB 33.4 + 0.6 KV | lossless |
| FP16 | 16 | 63.1 GB 62.5 + 0.6 KV | lossless |
Select your GPU above to see speed estimates and compatibility for each quantization.
Rent a GPU and run GLM 4.7 Flash now
Spin up an A100 / H100 / 4090 in ~60s. Pay by the second. Cancel anytime.
Detects your GPU, recommends the best model, downloads it, and starts chatting — zero config. Benchmarks your speed and contributes anonymous data to improve predictions.
At Q4_K_M quantization. Sorted by minimum VRAM.
Find the best GPU for GLM 4.7 Flash
Build Hardware for GLM 4.7 FlashGLM 4.7 Flash — fast inference variant for latency-sensitive applications.
Run GLM 4.7 Flash locally with Ollama — needs 19.4 GB VRAM at Q4_K_M:
ollama run glm-4.7-flash