Every open model in our directory, ranked by how well it runs on Reatan Mini Gaming PC (Ryzen AI 9 HX 470 with Speaker) (16GB) at 4-bit. Grades weigh whether the model fits in memory and how fast it should run.
Reatan Mini Gaming PC (Ryzen AI 9 HX 470 with Speaker) has 16 GB of VRAM, enough to run 113 of the 169 open models we track at 4-bit. The largest that still fits well is Mixtral 8x7B Instruct (46.7B), and the top-rated is LFM2.5-8B-A1B at about 25 tok/s.
113 of 169 models run comfortably on Reatan Mini Gaming PC (Ryzen AI 9 HX 470 with Speaker).
See full Reatan Mini Gaming PC (Ryzen AI 9 HX 470 with Speaker) specs and pricingThe top models this device can run at 4-bit, ranked by fit and speed.
| Model | Grade | Speed | VRAM |
|---|---|---|---|
| LFM2.5-8B-A1BLiquid AI | AA | 24.9 tok/s | 2.9 GB |
| Qwen3-30B-A3BAlibaba | BB | 13.5 tok/s | 5.4 GB |
| North Mini CodeCohere | BB | 8.6 tok/s | 8.4 GB |
| Nemotron 3 Nano OmniNVIDIA | BB | 8.5 tok/s | 8.5 GB |
| Qwen3.6 35B-A3BAlibaba |
LFM2.5-8B-A1B is the highest-rated model that runs on Reatan Mini Gaming PC (Ryzen AI 9 HX 470 with Speaker) at 4-bit, at about 25 tok/s. In total, Reatan Mini Gaming PC (Ryzen AI 9 HX 470 with Speaker) can run 113 of the 169 open models we track.
We size each model at 4-bit (Q4_K_M), add the KV cache for its context length and a fixed runtime overhead, then compare that to this device VRAM. Each model is graded on how comfortably it fits and how fast it should run on this card.
It is an estimate of decode speed in tokens per second at 4-bit, based on this device memory bandwidth and each model size. Real speed depends on your runtime, batch size, and context length, so use it to compare models, not as a hard promise.
Lower-bit quantization shrinks a model so more of them fit. These grades already assume 4-bit, which is the common local default. Going below 4-bit can squeeze a larger model in at some quality cost; going above needs more VRAM and may not fit.
| BB |
| 8.5 tok/s |
| 8.5 GB |
| Qwen3.5-35B-A3BAlibaba | BB | 8.5 tok/s | 8.5 GB |
| Mixtral 8x7B InstructMistral AI | BB | 6.4 tok/s | 11.4 GB |
| DiffusionGemma 26B-A4BGoogle | BB | 6.9 tok/s | 10.5 GB |
| Gemma 4 26B-A4B ITGoogle | BB | 6.6 tok/s | 11.0 GB |
| Llama 2 13B ChatMeta | BB | 8.6 tok/s | 8.5 GB |
| Llama 3 8B InstructMeta | BB | 12.8 tok/s | 5.7 GB |
| Carnice-9b for Hermes agentkai-os | BB | 12.0 tok/s | 6.0 GB |
LFM2.5-8B-A1BLiquid AI | 8.3B(1.5B active) | AA | 24.9 tok/s | 2.9 GB | |
Qwen3-30B-A3BAlibaba | 30B(3B active) | BB | 13.5 tok/s | 5.4 GB | |
North Mini CodeCohere | 30B(3B active) | BB | 8.6 tok/s | 8.4 GB | |
Nemotron 3 Nano OmniNVIDIA | 30B(3B active) | BB | 8.5 tok/s | 8.5 GB | |
Qwen3.6 35B-A3BAlibaba | 35B(3B active) | BB | 8.5 tok/s | 8.5 GB | |
Qwen3.5-35B-A3BAlibaba | 35B(3B active) | BB | 8.5 tok/s | 8.5 GB | |
Mixtral 8x7B InstructMistral AI | 46.7B(12.9B active) | BB | 6.4 tok/s | 11.4 GB | |
DiffusionGemma 26B-A4BGoogle | 25.2B(3.8B active) | BB | 6.9 tok/s | 10.5 GB | |
| Ad | |||||
Gemma 4 26B-A4B ITGoogle | 26B(4B active) | BB | 6.6 tok/s | 11.0 GB | |
Llama 2 13B ChatMeta | 13B | BB | 8.6 tok/s | 8.5 GB | |
| 8B | BB | 12.8 tok/s | 5.7 GB | ||
| 9B | BB | 12.0 tok/s | 6.0 GB | ||
PersonaPlex 7BNVIDIA | 7B | BB | 15.1 tok/s | 4.8 GB | |
Llama 2 7B ChatMeta | 7B | BB | 15.1 tok/s | 4.8 GB | |
Gemma 4 E4B ITGoogle | 4B | BB | 10.5 tok/s | 6.9 GB | |
Gemma 3 4B ITGoogle | 4B | BB | 10.5 tok/s | 6.9 GB | |
| Ad | |||||
Mistral 7B InstructMistral AI | 7B | BB | 11.3 tok/s | 6.4 GB | |
Gemma 4 E2B ITGoogle | 2B | BB | 19.5 tok/s | 3.7 GB | |
VibeThinker-3BWeiboAI | 3B | BB | 19.0 tok/s | 3.8 GB | |
| 8B | CC | 5.4 tok/s | 13.3 GB | ||
Qwen3.5-9BAlibaba | 9B | FF | 2.9 tok/s | 24.6 GB | |
Gemma 4 12BGoogle | 12B | FF | 2.3 tok/s | 32.0 GB | |
Gemma 4 12B Coderyuxinlu1 | 12B | FF | 2.3 tok/s | 32.0 GB | |
Mistral Small 3 24BMistral AI | 24B | FF | 1.9 tok/s | 39.0 GB | |
| Ad | |||||
Carnice-V2-27bkai-os | 27B | FF | 1.0 tok/s | 72.8 GB | |

Model Compatibility
Use the calculator to weigh this device against any other, model by model, with speed and fit scores.
Mixture-of-experts models only activate a fraction of their parameters at a time, so their runtime memory is closer to the active size than the headline size. We score fit on the active parameters, which is why some very large models still fit.