CogVLM2 19B needs ~27.7 GB VRAM. Gaudi 3 128GB has 128.0 GB. With Q4_K_M quantization, expect ~240 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
240.2 tok/s
TTFT
806 ms
Safe context
8K
Memory
27.7 GB / 128.0 GB
The raw memory story may look fine, but the software ecosystem is still a constraint here.
Runtime ecosystem is narrower than CUDA
Intel GPUs can look attractive on memory per dollar, but local AI tooling, kernels, and model coverage are still broader and easier on CUDA today.
Prefer CUDA if you want the path of least resistance
If your goal is maximum runtime coverage, easier troubleshooting, and better support for new local AI releases, CUDA is usually still the safer upgrade path.
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Chat | A | Runs well | 240.2 tok/s | 440 ms | 8K |
| Coding | A | Runs well | 240.2 tok/s | 806 ms | 8K |
| Agentic Coding | A | Runs well | 240.2 tok/s | 1172 ms | 8K |
| Reasoning | A | Runs well | 240.2 tok/s | 952 ms | 8K |
| RAG | A | Runs well | 240.2 tok/s | 1465 ms | 8K |
How CogVLM2 19B (19B params) fits at each quantization level on Gaudi 3 128GB (128.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 7.4 GB | Low | A71 |
Q3_K_S | 3 | 9.3 GB | Low | A72 |
NVFP4 | 4 | 10.6 GB | Medium | A72 |
Q4_K_M | 4 | 11.6 GB | Medium | A72 |
Q5_K_M | 5 | 13.7 GB | High | A72 |
Q6_K | 6 | 15.6 GB | High | A72 |
Q8_0 | 8 | 20.3 GB | Very High | A72 |
F16Best for your GPU | 16 | 38.9 GB | Maximum | A75 |
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 |
|---|---|---|---|---|
 Devstral 2 123B Instruct | 123B | S | 37.5 tok/s | |
 Qwen3-Coder 30B A3B Instruct | 30.5B | S | 391.6 tok/s | |
| Qwen 3.5 27B | 27B | S | 169.8 tok/s | |
| Qwen 3.6 27B | 27B | S | 105.9 tok/s | |
| Qwen 3.5 122B A10B | 122B | S | 104.1 tok/s |
Yes, Gaudi 3 128GB can run CogVLM2 19B with a A grade (Runs well). Expected decode speed: 240.2 tok/s.
CogVLM2 19B (19B parameters) requires approximately 27.7 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 Gaudi 3 128GB, CogVLM2 19B achieves approximately 240.2 tokens per second decode speed with a time-to-first-token of 806ms using Q4_K_M quantization.
For coding workloads, CogVLM2 19B on Gaudi 3 128GB receives a A grade with 240.2 tok/s and 8K context.
On Gaudi 3 128GB, 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.
Prefer CUDA if you want the path of least resistance. If your goal is maximum runtime coverage, easier troubleshooting, and better support for new local AI releases, CUDA is usually still the safer upgrade path.
Often yes, if your goal is the easiest setup and the widest runtime support. Intel can offer attractive memory capacity, but CUDA still tends to win on tooling maturity, guides, kernels, and model coverage for local AI.
Paste this snippet into any page to show a live fit card.
<iframe src="https://willitrunai.com/embed/cogvlm2-19b-on-gaudi-3-128gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>
Preview: