PC with RTX 4060, RTX 3070, RTX 5060, RTX 4060 Laptop or RTX 4070 Laptop and 32 GB RAM. Updated 2 November 2026. These machines fit exactly the same models, so they share this page. Their memory speeds range from 256 to 448 GB/s, so words per second scale with memory speed. Speeds shown are for RTX 4060.
PC with RTX 4060, RTX 3070, RTX 5060, RTX 4060 Laptop or RTX 4070 Laptop and 32 GB RAM can run 9 of the 31 models tracked here at a comfortable speed, and another 12 slowly. The largest chat or coding model that fits is Gemma 4 E4B, which needs about 7.4 GB and generates roughly 38 words a second. The quickest is Gemma 4 E2B at about 62 words a second, which feels faster than you can read.
Video memory (VRAM) is the limit: 8 GB on this card. A model must fit in VRAM to run fast. Anything larger spills into system RAM (32 GB in this setup), which works but is slow, because system RAM moves data at roughly 70 GB/s while the card moves 272 GB/s.
Best chat pick: Gemma 4 E4B at about 38 words a second. Best coding pick: Qwen2.5 Coder 7B at about 26.
Video models need more memory than this machine has.
Spare memory after loading the model decides how long a conversation or document can be before things slow down. With about 27 GB usable here:
The biggest single jump is Ryzen AI Max+ 395 (Strix Halo), 128 GB, which would unlock 18 more models, including Gemma 4 12B, gpt-oss 20B, Gemma 4 26B A4B.
For chat, Gemma 4 E4B is the best fit, at about 38 words a second. For coding, Qwen2.5 Coder 7B.
Gemma 4 12B runs at about 11 words a second (slow); gpt-oss 20B runs at about 11 words a second (slow); Qwen 3.5 35B A3B runs at about 11 words a second (slow).
No. Llama 3.3 70B needs about 45.6 GB and this machine tops out near 27 GB.
| Model | Verdict | Speed | Needs |
|---|---|---|---|
| Gemma 4 E4B | Runs great | 38 words/s | 7.4 GB |
| Qwen 3.5 9B | Runs great | 19 words/s | 7.4 GB |
| Qwen3 VL 8B | Runs great | 20 words/s | 6.9 GB |
| DeepSeek R1 8B | Runs great | 23 words/s | 6 GB |
| Qwen2.5 Coder 7B | Runs great | 26 words/s | 5.4 GB |
| Gemma 4 E2B | Runs great | 62 words/s | 5.3 GB |
| Qwen 3.5 4B | Runs great | 37 words/s | 4 GB |
| Whisper large-v3 turbo | Runs well | about 1 min | 4 GB |
| Kokoro TTS | Runs well | about 11 s | 2 GB |
| Qwen 3.6 35B A3B | Slow | 11 words/s | 24.6 GB |
| Qwen 3.5 35B A3B | Slow | 11 words/s | 23.6 GB |
| Qwen3 Coder 30B | Slow | 11 words/s | 20.4 GB |
| Gemma 4 26B A4B | Slow | 9 words/s | 19.4 GB |
| gpt-oss 20B | Slow | 11 words/s | 15.2 GB |
| Phi-4 14B | Slow | 8 words/s | 10.1 GB |
| Gemma 4 12B | Slow | 11 words/s | 8.9 GB |
| Stable Diffusion XL | Slow | about 36 s | 8 GB |
| Qwen3 TTS 1.7B | Slow | about 2 min | 8 GB |
| Gemma 4 31B | Very slow | 2 words/s | 21.5 GB |
| DeepSeek R1 32B | Very slow | 2 words/s | 21.5 GB |
| Qwen 3.6 27B | Very slow | 2 words/s | 19.9 GB |
As an Amazon Associate I earn from qualifying purchases.
Unlocks 14 more: Gemma 4 12B, gpt-oss 20B, Gemma 4 26B A4B, Gemma 4 31B and more.
Unlocks 17 more: Gemma 4 12B, gpt-oss 20B, Gemma 4 26B A4B, Qwen 3.5 35B A3B and more.
Unlocks 18 more: Gemma 4 12B, gpt-oss 20B, Gemma 4 26B A4B, Qwen 3.5 35B A3B and more.