Can the RTX 4070 run WeVisCut?
WeVisCut·9.4B·GeForce RTX 4070·12 GB VRAM
Yes. The RTX 4070 runs WeVisCut fully in VRAM at Q8_0 (about 11 GB of 12 GB), at roughly 28 tokens/s.
- best quant
- Q8_0
- needs
- 11 GB
- usable
- 12 GB
- est. speed
- ~28 tok/s
Estimates for an 8K context; offload assumes 32 GB of system RAM.Try other contexts on the WeVisCut page →
Every quantisation of WeVisCut on the RTX 4070
| Quant | Memory | Fit | Speed |
|---|---|---|---|
| Q2_K | 4.5 GB | fits | ~90 tok/s |
| Q3_K_M | 6.0 GB | fits | ~60 tok/s |
| Q4_K_M | 7.0 GB | fits | ~49 tok/s |
| Q5_K_M | 7.8 GB | fits | ~41 tok/s |
| Q6_K | 8.7 GB | fits | ~36 tok/s |
| Q8_0 pick | 11 GB | fits | ~28 tok/s |
| FP16 / BF16 | 19 GB | offload | ~3.9 tok/s |
Speed is a bandwidth-based estimate for generating text (504 GB/s); prompt processing and long contexts are slower.
Measured speed on the RTX 4070
No community reports for this pair yet.
Report your own speed →Run WeVisCut on the RTX 4070
Using Q8_0, the best fit for RTX 4070.
llama-server -hf prithivMLmods/tencent-WeVisCut-GGUF:Q8_0Downloads the GGUF on first run. Installed through llama.app? The same flags work with `llama serve`.
vllm serve tencent/WeVisCutservices:
vllm:
image: vllm/vllm-openai:latest
command: tencent/WeVisCut
ports:
- "8000:8000"
ipc: host
volumes:
- ~/.cache/huggingface:/root/.cache/huggingface
deploy:
resources:
reservations:
devices:
- driver: nvidia
count: all
capabilities: [gpu]API on port 8000.
sglang serve tencent/WeVisCut --host 0.0.0.0 --port 30000services:
sglang:
image: lmsysorg/sglang:latest
command: sglang serve tencent/WeVisCut --host 0.0.0.0 --port 30000
ports:
- "30000:30000"
ipc: host
volumes:
- ~/.cache/huggingface:/root/.cache/huggingface
deploy:
resources:
reservations:
devices:
- driver: nvidia
count: all
capabilities: [gpu]API on port 30000.
Omni models that fit the RTX 4070 comfortably
All that fit →| Model | Params | Best quant | Memory | Speed |
|---|---|---|---|---|
| 9.4B | Q8_0 | ≈10 GBFits in memory | ~28 tok/s | |
| 8B | Q8_0 | ≈9.2 GBFits in memory | ~33 tok/s | |
| 10.7B | Q8_0 | ≈8.9 GBFits in memory | ~24 tok/s | |
| 9.1B | Q8_0 | ≈10 GBFits in memory | ~29 tok/s | |
| 12B | Q4_K_M | ≈10 GBFits in memory | ~38 tok/s | |
| 8.3B | Q8_0 | ≈10 GBFits in memory | ~32 tok/s |
Other GPUs that run WeVisCut
Frequently asked questions
- Can the RTX 4070 run WeVisCut?
- Yes. The RTX 4070 runs WeVisCut fully in VRAM at Q8_0 (about 11 GB of 12 GB), at roughly 28 tokens/s.
- How fast does WeVisCut run on the RTX 4070?
- An estimate from memory bandwidth gives about 28 tokens/s at Q8_0.
- Which quantisation of WeVisCut should I use on the RTX 4070?
- Q8_0: the largest one that fits. Expected quality: lossless compared with the full-precision weights.
Memory and speed are estimates from the model's size and the card's memory bandwidth; real numbers depend on the runtime, drivers and context length. Community figures are medians of user reports.