32 GB VRAM (+8)1792 GB/s (+1492)
A
Raises estimated decode speed by about 808%.130.8 tok/s decode
Raises estimated decode speed by about 808%.
Adds memory headroom for longer context windows and future model growth.
~$1,999 MSRP
Can Nous Dolphin 13B run on NVIDIA L4 24GB?
YES — With Offload
B69Good
○Estimated from fit model
Nous Dolphin 13B needs ~25.2 GB VRAM. NVIDIA L4 24GB has 24.0 GB. With Q5_K_M quantization, expect ~14 tok/s.
Runtime: OllamaCapacity: OffloadBandwidth: LowStack: BasicBottleneck: Host offload
Share:
Operating mode
Choose the run profile you care about
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
Q5_K_M (High quality) — 25.2 GB, 14.4 tok/s, Runs with offload (needs ~0.4 GB host RAM)
25.2 GB required24.0 GB available
105% VRAM needed
1.2 GB over capacity — needs offload or smaller quantization
Fit status
Runs with offload (needs ~0.4 GB host RAM)
Decode
14.4 tok/s
TTFT
13424 ms
Safe context
14K
Memory
25.2 GB / 24.0 GB
Memory breakdown
Weights9.4 GB
KV Cache12.2 GB
Runtime1.2 GB
Headroom2.4 GB
See how fast it feels
See how fast it feelsNous Dolphin 13B on NVIDIA L4 24GB
1st promptCold start — includes initialization
>What is local AI inference?
Local AI inference means running an AI language model directly on your own hardware — your laptop, desktop, or server — instead of sending requests to a remote cloud API.
When you run inference locally the model weights are loaded into your GPU or unified memory. Each token you generate requires reading those weights from memory, so memory bandwidth is the main bottleneck for decode speed.
Key benefits of running locally:
- Full privacy: your prompts never leave your machine
- No per-token cost or rate limits
- Works offline once the model is downloaded
- Latency depends only on your hardware
2nd promptWarm — model ready, no init overhead
>How much VRAM do I need?
It depends on the model size and quantization level. A rough rule of thumb:
Model size Q4 (4-bit) Q8 (8-bit) FP16
7B params ~4.3 GB ~7.5 GB ~14 GB
13B params ~7.9 GB ~13.9 GB ~26 GB
70B params ~42.7 GB ~74.9 GB ~140 GB
Most people use 4-bit quantization (Q4_K_M) which gives 90-95% of full quality at a fraction of the memory. A 24 GB GPU can comfortably run most 7B-13B models.
Estimated: 14.4 tok/s decode · 13.4s TTFT (warm) · 36 tok/s prefill
What limits this setup
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.
Best improvement path
Buy headroom, not only minimum fit
A slightly larger memory tier gives you safer context growth and makes the recommendation more future-proof.
Performance by workload
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Chat | A | Runs well | 21.3 tok/s | 4969 ms | 14K |
| Coding | B | Runs with offload | 14.4 tok/s | 13424 ms | 14K |
| Agentic Coding | F | Too heavy | 6.3 tok/s | 44888 ms | 14K |
| Reasoning | B | Runs with offload (needs ~0.4 GB host RAM) | 14.4 tok/s | 15865 ms | 14K |
| RAG | F | Too heavy | 6.3 tok/s | 56110 ms | 14K |
Quantization options
How Nous Dolphin 13B (13B params) fits at each quantization level on NVIDIA L4 24GB (24.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 5.1 GB | Low | B66 |
Q3_K_S | 3 | 6.4 GB | Low | B67 |
NVFP4 | 4 |
Get started
Copy-paste commands to run Nous Dolphin 13B on your machine.
Run
docker run --rm -it ghcr.io/ggerganov/llama.cpp:full \
--hf-repo "nousresearch/Nous-Dolphin-13B" \
--hf-file "Nous-Dolphin-13B-Q5_K_M.gguf" \
-c 4096 -ngl 99Upgrade options
Hardware that runs Nous Dolphin 13B well
32 GB VRAM (+8)896 GB/s (+596)
A
Raises estimated decode speed by about 469%.82 tok/s decode
Raises estimated decode speed by about 469%.
Adds memory headroom for longer context windows and future model growth.
~$2,499 MSRP
32 GB VRAM (+8)576 GB/s (+276)
A
Raises estimated decode speed by about 249%.50.2 tok/s decode
Raises estimated decode speed by about 249%.
Adds memory headroom for longer context windows and future model growth.
~$4,000 MSRP