Only with CPU offload
Qwen3 14B at Q4_K_M needs 10.2 GB but only 7.0 GB is addressable, so about 38% of the layers would stream from system RAM at roughly 60 GB/s. Expect around 8.3 tokens per second — usable for batch work, painful for chat.
The largest Qwen3 that fits a 12 GB card at Q4 with room for context.
Hardware requirements What hardware do I need to run Qwen3 14B? →At Q4_K_M with an 8K context: a 16 GB GPU (RTX 5080) or a 24 GB Mac (M4 Pro). Minimum, recommended, high-quality and long-context tiers, with all 114 devices checked at every quantisation.
The VRAM budget
Quantisation ladder
| Quant | Weights | Total @ 8K | Max context | Tok/s | Quality | Fit |
|---|---|---|---|---|---|---|
| F16 | 27.6 GB | 29.4 GB | — | ~1.5 | Reference | 22.4 GB over |
| Q8_0 | 14.6 GB | 16.5 GB | — | ~3.4 | −0.1% ppl | 9.5 GB over |
| Q6_K | 11.3 GB | 13.2 GB | — | ~4.9 | −0.4% ppl | 6.2 GB over |
| Q5_K_M | 9.8 GB | 11.6 GB | — | ~6.2 | −0.8% ppl | 4.6 GB over |
| Q4_K_M | 8.3 GB | 10.2 GB | — | ~8.3 | −1.9% ppl | 3.2 GB over |
| Q3_K_M | 6.7 GB | 8.6 GB | — | ~13 | −5.4% ppl | 1.6 GB over |
| Q2_K | 5.8 GB | 7.6 GB | 4K | ~20 | −15% ppl | 0.6 GB over |
Quality is the published perplexity delta against f16 weights. Max context assumes an f16 KV cache; q8_0 roughly doubles it.
How to run it
$ llama-server \
-hf Qwen/Qwen3-14B:Q4_K_M \
-c 8192 -ngl 24
The engine underneath most of the others. Every knob is exposed. More on llama.cpp.
Related models on a GeForce RTX 4070 Laptop
| Model | Total | Tok/s est. | |
|---|---|---|---|
| Qwen3 8B 8.19B · same family |
Q4_K_M · 6.3 GB | 34 | Tight fit |
| Qwen3 30B-A3B 30.5B MoE · same family |
Q4_K_M · 18.5 GB | ~10 | CPU offload |
| Qwen3 Coder 30B-A3B 30.5B MoE · same family |
Q4_K_M · 18.5 GB | ~10 | CPU offload |
| Qwen3 32B 32.8B · same family |
Q4_K_M · 21.0 GB | ~2.4 | CPU offload |
| DeepSeek-R1-Distill-Qwen 14B 14.8B · similar size |
Q4_K_M · 10.4 GB | ~7.9 | CPU offload |
| Qwen2.5-Coder 14B 14.8B · similar size |
Q4_K_M · 10.4 GB | ~7.9 | CPU offload |
| Phi-4 14B 14.7B · similar size |
Q4_K_M · 10.4 GB | ~8.0 | CPU offload |
Qwen3 14B on other hardware
| Device | Total | Tok/s est. | |
|---|---|---|---|
| GeForce RTX 4060 Laptop 7.0 GB usable · 256 GB/s |
10.2 GB | ~8.3 | CPU offload |
| GeForce RTX 3060 12 GB 10.6 GB usable · 360 GB/s |
10.2 GB | 26 | Tight fit |
| GeForce RTX 5080 14.4 GB usable · 960 GB/s |
10.2 GB | 70 | Runs great |
| M4 · 16 GB 10.7 GB usable · 120 GB/s |
10.2 GB | 8.1 | Tight fit |
| GeForce RTX 4090 22.4 GB usable · 1008 GB/s |
10.2 GB | 73 | Runs great |
| Radeon RX 7900 XTX 22.4 GB usable · 960 GB/s |
10.2 GB | 70 | Runs great |
| M4 Pro · 24 GB 18.0 GB usable · 273 GB/s |
10.2 GB | 18 | Runs great |
| M3 Max · 64 GB 48.0 GB usable · 400 GB/s |
10.2 GB | 27 | Runs great |