Skip to content

What can the Radeon AI PRO R9700 run?

What fits on the Radeon AI PRO R9700
Text models: 233Visual models: 15Strongest first
Memory
32 GB
Bandwidth
640 GB/s
Maker
AMD

Open the calculator on this machineEvery model for 32 GB of VRAM

Text models that run on it

Text models that run on the Radeon AI PRO R9700, best first
ModelBest fitContextSpeedStrongest at
Qwen3.8 27B8-bit24K13 tok/sChat, Coding, Agents, Docs
Muse Glimmer 30B6-bit128K16 tok/sDocs
Qwen3.6 35B A3B6-bit96K98 tok/sChat, Coding, Agents, Docs
Gemma 4 31B6-bit112K14 tok/sChat, Coding, Agents, Docs
Gemma 4 26B A4B8-bit256K76 tok/sChat, Coding, Agents, Docs
Qwen3.5 9B8-bit256K39 tok/sChat, Coding, Agents, Docs
GLM-4.7 Flash6-bit104K94 tok/sChat, Coding, Agents
Gemma 4 12B8-bit256K28 tok/sChat, Coding, Docs
Nemotron 3.5 Lightning 30B A3B6-bit256K89 tok/sAgents, Docs
Devstral Small 2 24BFP832K15 tok/sCoding
LFM2.5 24B A2B8-bit128K109 tok/sAgents
Mistral Small 3.2 24B8-bit36K15 tok/sChat, Coding, Agents
Ministral 3 14B8-bit96K26 tok/sCoding, Agents, Docs
Gemma 4 E4B8-bit128K61 tok/sChat, Coding, Docs
gpt-oss 20BMXFP4128K113 tok/sAgents
Ministral 3 8B8-bit152K40 tok/sDocs
Olmo 3.1 32B6-bit64K14 tok/sChat, Coding, Agents, Docs
Llama 3.1 8B8-bit128K45 tok/sChat, Coding, Agents, Docs
Qwen3 14B8-bit40K25 tok/sNot rated
Qwen3 VL 8B Instruct8-bit144K41 tok/sText and images
Qwen3 8B8-bit40K44 tok/sNot rated
Qwen2.5 7B Instruct8-bit32K48 tok/sNot rated
Qwen2.5 VL 7B Instruct8-bit125K44 tok/sText and images
Qwen3 VL 4B Instruct8-bit176K71 tok/sText and images
Qwen3 4B8-bit40K77 tok/sNot rated
Qwen3 32B6-bit16K14 tok/sNot rated
Qwen3 4B Instruct 25078-bit176K77 tok/sNot rated
Qwen3.6 27B8-bit24K13 tok/sText and images
Qwen2.5 Coder 14B Instruct8-bit32K24 tok/sCoding
Qwen3 Coder 30B A3B Instruct6-bit64K93 tok/sCoding
Qwen3.5 27B8-bit24K13 tok/sText and images
Qwen3.5 35B A3B6-bit96K98 tok/sText and images
NVIDIA Nemotron 3 Nano 30B A3B BF166-bit256K89 tok/sNot rated
Qwen2.5 Coder 7B Instruct8-bit32K48 tok/sCoding
Mistral 7B Instruct V0.28-bit32K47 tok/sNot rated
Qwen3 30B A3B6-bit40K93 tok/sNot rated
Qwen3 30B A3B Instruct 25076-bit64K93 tok/sNot rated
Laguna XS.26-bit104K100 tok/sNot rated
OTel 2.0 LLM 31B IT6-bit112K14 tok/sNot rated
Qwen2.5 32B Instruct6-bit16K14 tok/sNot rated
MiniCPM V 4.7 35B A3B6-bit120K98 tok/sText and images
Qwen2.5 14B Instruct8-bit32K24 tok/sNot rated
Laguna XS 2.16-bit104K100 tok/sNot rated
Diffusiongemma 26B A4B IT8-bit184K73 tok/sText and images
Qwen2.5 VL 32B Instruct5-bit28K16 tok/sText and images
Qwen2.5 Coder 32B Instruct6-bit16K14 tok/sCoding
Qwen2 VL 7B Instruct8-bit32K44 tok/sText and images
Ornith 1.5 9B8-bit256K39 tok/sNot rated
Qwen3 VL 30B A3B Instruct6-bit60K93 tok/sText and images
Mistral 7B V0.18-bit32K47 tok/sNot rated
Meta Llama 3 8B Instruct Abliterated V38-bit8K45 tok/sNot rated
EXAONE 3.5 7.8B Instruct8-bit32K45 tok/sNot rated
Saiga Llama3 8B8-bit8K45 tok/sNot rated
DeepSeek R1 0528 Qwen3 8B8-bit128K44 tok/sNot rated
Gemma 3 12B IT8-bit128K30 tok/sText and images
Phi 3.5 Mini Instruct8-bit64K84 tok/sNot rated
DeepSeek R1 Distill Qwen 32B6-bit16K14 tok/sNot rated
Qwen3 4B Thinking 25078-bit176K77 tok/sNot rated
Gemma 3 27B IT8-bit20K13 tok/sText and images
Phi 4 Mini Instruct8-bit128K82 tok/sNot 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 233 are shown.

Image and video models

Image and video models that run on the Radeon AI PRO R9700
ModelBest fitPer outputStrongest at
MiniMax H38-bit8.9 minVideo
LTX-2.58-bit71 sVideo
FLUX.1 [dev]16-bit17 sImages
Qwen-Image8-bit5.1 minImages
FLUX.2 [dev]6-bit2.6 minImages
Qwen-Image 2.116-bit2.2 minImages
Wan2.2 I2V A14B8-bit46 minVideo
FLUX.2 [klein] 9B16-bit4.7 sImages
Wan2.1 I2V 14B8-bit46 minVideo
Wan 2.1 T2V-14B8-bit57 minVideo
Krea 2 Turbo16-bit17 sImages
Stable Diffusion 3.5 Large16-bit70 sImages
Qwen Image 25128-bit5.1 minImages
Wan 2.2 A14B8-bit46 minVideo
HunyuanVideo 1.516-bit56 minVideo

Only just too big

Questions

What is the best LLM for the Radeon AI PRO R9700?

Of the open models we track, the strongest that run well on the Radeon AI PRO R9700 are Qwen3.8 27B, Muse Glimmer 30B, Qwen3.6 35B A3B. 233 text models fit it in all.

Calculated from published specs, not benchmarked. All machines