What can the Ryzen AI Max+ 395 128 GB run?
| Text models: 119 | Visual models: 9 | Strongest first |
- Memory
- 128 GB unified
- Bandwidth
- 256 GB/s
- Maker
- AMD
- Also called
- Strix Halo
Open the calculator on this machineRyzen AI Max+ 395 (Strix Halo) in our inventory
Text models that run on it
| Model | Best fit | Context | Speed | Strongest at |
|---|---|---|---|---|
| Qwen3.8 Flash-Next | 3-bit | 256K | 41 tok/s | Chat, Coding, Agents |
| MiniMax M3 | 1.6-bit | 64K | 28 tok/s | Docs |
| Qwen3.8 27B | 8-bit | 256K | 5.3 tok/s | Chat, Coding, Agents, Docs |
| Qwen3.5 397B A17B | 1.6-bit | 256K | 34 tok/s | Agents, Docs |
| MiniMax M2.7 | 2-bit | 40K | 33 tok/s | Coding, Agents |
| Muse Glimmer 30B | 8-bit | 128K | 5.0 tok/s | Docs |
| Mistral Medium 3.5 128B | 5-bit | 16K | 1.7 tok/s | Chat, Coding, Agents, Docs |
| Qwen3.5 122B A10B | 5-bit | 256K | 19 tok/s | Chat, Coding, Agents, Docs |
| Nemotron 3 Super 120B A12B | 5-bit | 256K | 15 tok/s | Chat, Coding, Agents, Docs |
| Nemotron 3.5 Bolt 74B A9B | 8-bit | 256K | 14 tok/s | Chat, Coding, Agents, Docs |
| Qwen3.6 35B A3B | 8-bit | 256K | 40 tok/s | Chat, Coding, Agents, Docs |
| Gemma 4 31B | 8-bit | 256K | 4.6 tok/s | Chat, Coding, Agents, Docs |
| Gemma 4 26B A4B | 8-bit | 256K | 34 tok/s | Chat, Coding, Agents, Docs |
| GLM-4.7 Flash | 8-bit | 198K | 40 tok/s | Chat, Coding, Agents |
| Mistral Small 4 119B | 5-bit | 256K | 30 tok/s | Chat, Docs |
| gpt-oss 120B | MXFP4 | 128K | 37 tok/s | Chat, Coding |
| Qwen3 Coder Next 80B | 8-bit | 256K | 39 tok/s | Coding, Agents |
| Nemotron 3.5 Lightning 30B A3B | 8-bit | 256K | 39 tok/s | Agents, Docs |
| Devstral Small 2 24B | FP8 | 384K | 6.1 tok/s | Coding |
| LFM2.5 24B A2B | 8-bit | 128K | 57 tok/s | Agents |
| Mistral Small 3.2 24B | 8-bit | 128K | 6.1 tok/s | Chat, Coding, Agents |
| Llama 3.3 70B | 8-bit | 64K | 2.1 tok/s | Chat, Coding, Agents, Docs |
| Olmo 3.1 32B | 8-bit | 64K | 4.5 tok/s | Chat, Coding, Agents, Docs |
| Qwen3 32B | 8-bit | 40K | 4.5 tok/s | Not rated |
| Qwen3.6 27B | 8-bit | 256K | 5.3 tok/s | Text and images |
| Qwen3 Coder 30B A3B Instruct | 8-bit | 256K | 40 tok/s | Coding |
| Qwen3.5 27B | 8-bit | 256K | 5.3 tok/s | Text and images |
| Qwen3.5 35B A3B | 8-bit | 256K | 40 tok/s | Text and images |
| NVIDIA Nemotron 3 Nano 30B A3B BF16 | 8-bit | 256K | 39 tok/s | Not rated |
| GLM 4.5 Air | 6-bit | 32K | 12 tok/s | Not rated |
| Inkling Small | 2-bit | 4K | 32 tok/s | Text and images |
| Qwen3 30B A3B | 8-bit | 40K | 40 tok/s | Not rated |
| MiniMax M2.5 | 2-bit | 40K | 30 tok/s | Not rated |
| Qwen3 30B A3B Instruct 2507 | 8-bit | 256K | 40 tok/s | Not rated |
| Laguna XS.2 | 8-bit | 256K | 41 tok/s | Not rated |
| OTel 2.0 LLM 31B IT | 8-bit | 256K | 4.6 tok/s | Not rated |
| Laguna M.1 | 2-bit | 40K | 16 tok/s | Not rated |
| Qwen2.5 32B Instruct | 8-bit | 32K | 4.5 tok/s | Not rated |
| MiniCPM V 4.7 35B A3B | 8-bit | 256K | 40 tok/s | Text and images |
| Laguna XS 2.1 | 8-bit | 256K | 41 tok/s | Not rated |
| Diffusiongemma 26B A4B IT | 8-bit | 256K | 33 tok/s | Text and images |
| Qwen2.5 VL 32B Instruct | 8-bit | 125K | 4.4 tok/s | Text and images |
| Qwen2.5 Coder 32B Instruct | 8-bit | 32K | 4.5 tok/s | Coding |
| Laguna S 2.1 | 6-bit | 16K | 20 tok/s | Not rated |
| Qwen3 VL 30B A3B Instruct | 8-bit | 256K | 40 tok/s | Text and images |
| DeepSeek R1 Distill Qwen 32B | 8-bit | 128K | 4.5 tok/s | Not rated |
| Gemma 3 27B IT | 8-bit | 128K | 5.5 tok/s | Text and images |
| Qwen3 VL 235B A22B Instruct | 2-bit | 44K | 16 tok/s | Text and images |
| Qwen3 235B A22B | 2-bit | 40K | 16 tok/s | Not rated |
| Ornith 1.5 35B A3B | 8-bit | 256K | 40 tok/s | Not rated |
| Ornith 1.5 122B | 5-bit | 256K | 16 tok/s | Not rated |
| Ornith 1.5 397B | 1.6-bit | 256K | 25 tok/s | Not rated |
| Qwen3 VL 32B Instruct | 8-bit | 224K | 4.4 tok/s | Text and images |
| Step 3.5 Flash | 2-bit | 52K | 37 tok/s | Not rated |
| Qwen2.5 72B Instruct | 8-bit | 32K | 2.0 tok/s | Not rated |
| Qwen3 Next 80B A3B Instruct | 8-bit | 256K | 39 tok/s | Not rated |
| Llama 4 Scout 17B 16E Instruct | 6-bit | 36K | 9.9 tok/s | Text and images |
| MiniMax M2 | 2-bit | 40K | 30 tok/s | Not rated |
| Qwen2.5 VL 72B Instruct | 8-bit | 56K | 2.0 tok/s | Text and images |
| Qwen1.5 7B | 8-bit | 32K | 20 tok/s | Not rated |
Best fit is the highest quality, up to 8-bit, that leaves the use case’s context. Ranked by our use-case scores, then by Hugging Face downloads. Speeds are rough, plus or minus 25%. The top 60 of 119 are shown.
Image and video models
| Model | Best fit | Per output | Strongest at |
|---|---|---|---|
| MiniMax H3 | 16-bit | 28 min | Video |
| LTX-2.5 | 16-bit | 3.1 min | Video |
| Qwen-Image | 16-bit | 17 min | Images |
| FLUX.2 [dev] | 16-bit | 7.3 min | Images |
| Wan2.2 I2V A14B | 16-bit | 2.4 h | Video |
| Wan2.1 I2V 14B | 16-bit | 2.4 h | Video |
| Wan 2.1 T2V-14B | 16-bit | 3.1 h | Video |
| Qwen Image 2512 | 16-bit | 17 min | Images |
| Wan 2.2 A14B | 16-bit | 2.4 h | Video |
Only just too big
- GLM-5.3 Flash
- MiMo V2.6 Flash
- DeepSeek V4 Flash
- Hy3
- Nemotron 3 Ultra 550B A55B
- MiMo V2.5
- GLM 5.3 Flash BF16
- Hy3 Preview
- Qwen3 Coder 480B A35B Instruct
- MiMo V2 Flash
- Beam 1.0 501B A23B
These need up to a quarter more memory than the Ryzen AI Max+ 395 128 GB has. Spilling into system RAM or a lower precision can get them running, slowly.
Questions
What is the best LLM for the Ryzen AI Max+ 395 128 GB?
Of the open models we track, the strongest that run well on the Ryzen AI Max+ 395 128 GB are Qwen3.8 Flash-Next, MiniMax M3, Qwen3.8 27B. 119 text models fit it in all.
Calculated from published specs, not benchmarked. All machines