Can the GeForce RTX 5060 Ti 16GB run GLM-4.7-Flash?
Only with CPU offload at Q4_K_M.
GLM-4.7-Flash is 30B parameters (MoE; all experts resident). At Q4_K_M the weights are about 16.8 GiB; with the 1.5 GiB runtime reserve that is 18.3 GiB against the card's 16GB, about 2.3 GiB more than the card has.
All three quants
16GB card| Quant | Weights | With reserve | Verdict | Headroom |
|---|---|---|---|---|
| Q4_K_Mthe everyday quant | 16.8 GiB | 18.3 GiB | Only with CPU offload | short 2.3 GiB |
| Q8_0near-lossless | 29.6 GiB | 31.1 GiB | Only with CPU offload | short 15.1 GiB |
| FP16full weights | 55.9 GiB | 57.4 GiB | Does not fit | short 41.4 GiB |
Verdict rules: fits comfortably when weights plus reserve sit within 85% of VRAM; tight when they fit with little left for context; CPU offload when they exceed VRAM by up to 2x (runs, slowly, with layers in system RAM); no beyond that. The KV cache per token depends on the model's architecture and is not modelled here.
Not this card. These fit it comfortably at Q4.
- GeForce RTX 5090 32GB · headroom 13.7 GiB
- GeForce RTX 3090 (used) 24GB · headroom 5.7 GiB
- GeForce RTX 4090 (used) 24GB · headroom 5.7 GiB
- Radeon RX 7900 XTX 24GB · headroom 5.7 GiB
What the GeForce RTX 5060 Ti 16GB does run well → · GLM-4.7-Flash on every card →
About the model
GLM-4.7-Flash by Z.ai (Zhipu): 202,752-token native context, text, released 2026 under the MIT. Model card → · Hosted pricing on PicksByModel →