CogVLM2 19B needs ~16.5 GB VRAM. Radeon RX 7900M 16GB has 16.0 GB. With Q4_K_M quantization, expect ~22 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
0.5 GB over capacity — needs offload or smaller quantization
Fit status
Runs with offload (needs ~0.4 GB host RAM)
Decode
22.1 tok/s
TTFT
8772 ms
Safe context
8K
Memory
16.5 GB / 16.0 GB
This setup is broadly balanced for this model.
Very little memory headroom
You can run the model, but there is not much room left for longer context, bigger batches, extra apps, or future model updates.
Buy headroom, not only minimum fit
A slightly larger memory tier gives you safer context growth and makes the recommendation more future-proof.
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Chat | A | Runs with offload | 31.5 tok/s | 3350 ms | 8K |
| Coding | A | Runs with offload (needs ~0.4 GB host RAM) | 22.1 tok/s | 8772 ms | 8K |
| Agentic Coding | B | Very compromised (needs ~1.8 GB host RAM) | 16.5 tok/s | 17052 ms | 8K |
| Reasoning | A | Runs with offload (needs ~0.4 GB host RAM) | 22.1 tok/s | 10367 ms | 8K |
| RAG | B | Very compromised (needs ~1.8 GB host RAM) | 16.5 tok/s | 21315 ms | 8K |
How CogVLM2 19B (19B params) fits at each quantization level on Radeon RX 7900M 16GB (16.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 7.4 GB | Low | A85 |
Q3_K_S | 3 | 9.3 GB | Low | A84 |
NVFP4 | 4 | 10.6 GB | Medium | A84 |
Q4_K_MBest for your GPU | 4 | 11.6 GB | Medium | A84 |
Q5_K_M | 5 | 13.7 GB | High | F0 |
Q6_K | 6 | 15.6 GB | High | F0 |
Q8_0 | 8 | 20.3 GB | Very High | F0 |
F16 | 16 | 38.9 GB | Maximum | F0 |
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 |
|---|---|---|---|---|
 GPT-OSS 20B | 21B | A | 39.3 tok/s | |
 Codestral 2 25.08 | 22B | A | 14.4 tok/s |
Yes, Radeon RX 7900M 16GB can run CogVLM2 19B with a A grade (Runs with offload (needs ~0.4 GB host RAM)). Expected decode speed: 22.1 tok/s.
CogVLM2 19B (19B parameters) requires approximately 16.5 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 RX 7900M 16GB, CogVLM2 19B achieves approximately 22.1 tokens per second decode speed with a time-to-first-token of 8772ms using Q4_K_M quantization.
For coding workloads, CogVLM2 19B on Radeon RX 7900M 16GB receives a A grade with 22.1 tok/s and 8K context.
On Radeon RX 7900M 16GB, 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.
Buy headroom, not only minimum fit. A slightly larger memory tier gives you safer context growth and makes the recommendation more future-proof.
Paste this snippet into any page to show a live fit card.
<iframe src="https://willitrunai.com/embed/cogvlm2-19b-on-rx-7900m-16gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>
Preview: