GLM-4 9B Chat
9.4B parameters · chatglm
Estimated VRAM requirement
Weight memory uses parameter count × bits per weight, plus a quantization overhead factor. KV cache uses standard GQA formulas. Values are calculated estimates, not measured allocations.
| Quantization | Estimated weights | 2K total | 4K total | 8K total | 16K total | 32K total | 64K total | 128K total |
|---|---|---|---|---|---|---|---|---|
| FP16 | 17.9 GB | 21.3 GB | 22.5 GB | 25.0 GB | 30.0 GB | 40.0 GB | 60.0 GB | 100.0 GB |
| Q8 | 10.0 GB | 12.8 GB | 14.1 GB | 16.6 GB | 21.6 GB | 31.6 GB | 51.6 GB | 91.6 GB |
| Q6 | 7.8 GB | 10.5 GB | 11.7 GB | 14.2 GB | 19.2 GB | 29.2 GB | 49.2 GB | 89.2 GB |
| Q5 | 6.6 GB | 9.2 GB | 10.4 GB | 12.9 GB | 17.9 GB | 27.9 GB | 47.9 GB | 87.9 GB |
| Q4 | 5.5 GB | 8.0 GB | 9.2 GB | 11.7 GB | 16.7 GB | 26.7 GB | 46.7 GB | 86.7 GB |
| Q3 | 4.4 GB | 6.8 GB | 8.0 GB | 10.5 GB | 15.5 GB | 25.5 GB | 45.5 GB | 85.5 GB |
Find a GPU for this model
Run at the quantization and context selected below. Defaults are Q4 and 8K when you do not change the controls. Results are full-VRAM fits only — not a speed ranking.
- Run at
- Q4 · 8K context
- Required VRAM
- 11.7 GB
No fresh matched Amazon price is currently available for a full-VRAM fit. See compatible GPUs · Workstations · Rent a server · Buy vs rent.
Lowest-cost full-VRAM option
Sorted by lowest current Amazon price among GPUs that fit entirely in VRAM. This is not a performance ranking.
No compatible GPU currently has a fresh Amazon price.
More VRAM headroom
Sorted by leftover usable VRAM after the calculated requirement. Headroom is not tokens/sec.
NVIDIA B200 192 GB
- GPU VRAM
- 192 GB advertised
- Required VRAM
- 11.7 GB
- Usable GPU VRAM
- 172.8 GB
- VRAM headroom
- 161.1 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 0 mapped · 0 Amazon-matched
- Amazon
- Check availability
NVIDIA H200 141 GB
- GPU VRAM
- 141 GB advertised
- Required VRAM
- 11.7 GB
- Usable GPU VRAM
- 126.9 GB
- VRAM headroom
- 115.2 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 0 mapped · 0 Amazon-matched
- Amazon
- Check availability
NVIDIA GH200 96 GB
- GPU VRAM
- 96 GB advertised
- Required VRAM
- 11.7 GB
- Usable GPU VRAM
- 86.4 GB
- VRAM headroom
- 74.7 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 0 mapped · 0 Amazon-matched
- Amazon
- Check availability
Other compatible GPUs
NVIDIA RTX 4500 24 GB
- GPU VRAM
- 24 GB advertised
- Required VRAM
- 11.7 GB
- Usable GPU VRAM
- 21.6 GB
- VRAM headroom
- 9.9 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 2 mapped · 0 Amazon-matched
- Amazon
- Check availability
NVIDIA RTX 3090 24 GB
- GPU VRAM
- 24 GB advertised
- Required VRAM
- 11.7 GB
- Usable GPU VRAM
- 21.6 GB
- VRAM headroom
- 9.9 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 18 mapped · 9 Amazon-matched
- Amazon
- Check price on Amazon
NVIDIA RTX 5000 32 GB
- GPU VRAM
- 32 GB advertised
- Required VRAM
- 11.7 GB
- Usable GPU VRAM
- 28.8 GB
- VRAM headroom
- 17.1 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 2 mapped · 1 Amazon-matched
- Amazon
- Check price on Amazon
Compatible GPUs
Status is a calculated memory fit against usable VRAM (advertised × 0.9). Open a GPU to see product SKUs and Amazon listings.
Compare GPUs for this model
Published chip comparisons where both GPUs fully fit this model at Q4 / 8K. Opens with this model preselected; that query is not indexed.
How fast can this model run?
llama.cpp CUDA decode at Q4 under llama-bench tg128 conditions. Not a ranking. Open a pair for prefill and methodology.
| GPU | Decode | State | Confidence |
|---|---|---|---|
| NVIDIA RTX 5090 32 GB | ~194 tok/s | Calibrated estimate | MEDIUM |
| NVIDIA RTX 4090 24 GB | ~122 tok/s | Calibrated estimate | MEDIUM |
| NVIDIA RTX 5080 16 GB | ~119 tok/s | Calibrated estimate | MEDIUM |
| NVIDIA RTX 5070 Ti 16 GB | ~118 tok/s | Calibrated estimate | MEDIUM |
| NVIDIA RTX 3090 Ti 24 GB | ~111 tok/s | Calibrated estimate | MEDIUM |
| NVIDIA RTX 3090 24 GB | ~104 tok/s | Calibrated estimate | MEDIUM |
| NVIDIA RTX 4080 SUPER 16 GB | ~95 tok/s | Calibrated estimate | MEDIUM |
| NVIDIA RTX A6000 48 GB | ~94 tok/s | Calibrated estimate | MEDIUM |
| NVIDIA RTX 4080 16 GB | ~93 tok/s | Calibrated estimate | MEDIUM |
| NVIDIA RTX 4070 Ti SUPER 16 GB | ~86 tok/s | Calibrated estimate | MEDIUM |
Sources
Model specifications: official repository THUDM/glm-4-9b-chat · revision main. Compatibility: RigForAI calculator v1.0.0 (estimated VRAM). GPU products: Icecat. Amazon is a separate affiliate match and is not used as a spec source.