CogVLM2 19B needs ~19.7 GB VRAM. Radeon PRO W7900 DS 48GB has 48.0 GB. With Q4_K_M quantization, expect ~47 tok/s.
Operating mode
Interactive favors responsiveness, while light API and scale-out lean harder on serving readiness. The fit stays the same, but the recommendation lens changes.
Current mode
Balanced
Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.
Select quantization to explore
Fit status
Runs well
Decode
47.3 tok/s
TTFT
4095 ms
Safe context
8K
Memory
19.7 GB / 48.0 GB
This setup is broadly balanced for this model.
No major red flags
This recommendation has enough memory headroom and acceptable estimated speed for the selected workload.
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Chat | A | Runs well | 44.0 tok/s | 2401 ms | 8K |
| Coding | A | Runs well | 47.3 tok/s | 4095 ms | 8K |
| Agentic Coding | A | Runs well | 47.3 tok/s | 5956 ms | 8K |
| Reasoning | A | Runs well | 47.3 tok/s | 4839 ms | 8K |
| RAG | A | Runs well | 47.3 tok/s | 7445 ms | 8K |
How CogVLM2 19B (19B params) fits at each quantization level on Radeon PRO W7900 DS 48GB (48.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 7.4 GB | Low | A75 |
Q3_K_S | 3 | 9.3 GB | Low | A76 |
NVFP4 | 4 |
Copy-paste commands to run CogVLM2 19B on your machine.
Run
docker run --rm -it ghcr.io/ggerganov/llama.cpp:full \
--hf-repo "THUDM/cogvlm2-llama3-chat-19B" \
--hf-file "cogvlm2-llama3-chat-19B-Q4_K_M.gguf" \
-c 4096 -ngl 99Your hardware
| Model | Params | Grade | Decode | Capabilities |
|---|---|---|---|---|
| 30.5B | S | 77.1 tok/s | ||
| 27B | S | 33.4 tok/s |
Yes, Radeon PRO W7900 DS 48GB can run CogVLM2 19B with a A grade (Runs well). Expected decode speed: 47.3 tok/s.
CogVLM2 19B (19B parameters) requires approximately 19.7 GB of memory with Q4_K_M quantization.
The recommended quantization for CogVLM2 19B is Q4_K_M, which balances quality and memory efficiency.
On Radeon PRO W7900 DS 48GB, CogVLM2 19B achieves approximately 47.3 tokens per second decode speed with a time-to-first-token of 4095ms using Q4_K_M quantization.
For coding workloads, CogVLM2 19B on Radeon PRO W7900 DS 48GB receives a A grade with 47.3 tok/s and 8K context.
On Radeon PRO W7900 DS 48GB, CogVLM2 19B can safely use up to 8K tokens of context. The model's official context limit is 8K, but available memory constrains the safe maximum.
Paste this snippet into any page to show a live fit card.
<iframe src="https://willitrunai.com/embed/cogvlm2-19b-on-radeon-pro-w7900-ds-48gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>
Preview:
10.6 GB |
| Medium |
| A76 |
Q4_K_M | 4 | 11.6 GB | Medium | A77 |
Q5_K_M | 5 | 13.7 GB | High | A77 |
Q6_K | 6 | 15.6 GB | High | A78 |
Q8_0 | 8 | 20.3 GB | Very High | A79 |
F16Best for your GPU | 16 | 38.9 GB | Maximum | A81 |
| 27B | S | 23.9 tok/s |
| 35B | S | 64.8 tok/s |
| 30B | S | 79.7 tok/s |