Skip to content

What can the Radeon RX 7900 XTX run?

What fits on the Radeon RX 7900 XTX
Text models: 258Visual models: 17Strongest first
Memory
24 GB
Bandwidth
960 GB/s
Maker
AMD

Open the calculator on this machineEvery model for 24 GB of VRAM

Text models that run on it

Text models that run on the Radeon RX 7900 XTX, best first
ModelBest fitContextSpeedStrongest at
Qwen3.8 27B5-bit48K28 tok/sChat, Coding, Agents, Docs
Muse Glimmer 30B5-bit128K27 tok/sDocs
Qwen3.6 35B A3B4-bit72K146 tok/sChat, Coding, Agents, Docs
Gemma 4 31B4-bit80K28 tok/sChat, Coding, Agents, Docs
Gemma 4 26B A4B6-bit184K123 tok/sChat, Coding, Agents, Docs
Qwen3.5 9B8-bit256K57 tok/sChat, Coding, Agents, Docs
GLM-4.7 Flash5-bit24K126 tok/sChat, Coding, Agents
Gemma 4 12B8-bit256K41 tok/sChat, Coding, Docs
Nemotron 3.5 Lightning 30B A3B5-bit176K118 tok/sAgents, Docs
Qwen3.5 4B8-bit256K97 tok/sAgents, Docs
Devstral Small 2 24B6-bit20K28 tok/sCoding
LFM2.5 24B A2B6-bit128K155 tok/sAgents
Mistral Small 3.2 24B6-bit20K28 tok/sChat, Coding, Agents
Ministral 3 14B8-bit48K38 tok/sCoding, Agents, Docs
Gemma 4 E4B8-bit128K86 tok/sChat, Coding, Docs
gpt-oss 20BMXFP4128K151 tok/sAgents
Ministral 3 8B8-bit96K58 tok/sDocs
Gemma 4 E2B8-bit128K146 tok/sChat
Olmo 3.1 32B4-bit48K27 tok/sChat, Coding, Agents, Docs
Llama 3.1 8B8-bit112K65 tok/sChat, Coding, Agents, Docs
Qwen3 14B8-bit40K36 tok/sNot rated
Qwen3 VL 8B Instruct8-bit88K59 tok/sText and images
Qwen3 8B8-bit40K63 tok/sNot rated
Qwen2.5 7B Instruct8-bit32K69 tok/sNot rated
Qwen2.5 VL 7B Instruct8-bit125K64 tok/sText and images
Qwen3 VL 4B Instruct8-bit120K100 tok/sText and images
Qwen3 1.7B8-bit40K194 tok/sNot rated
Qwen3 4B8-bit40K108 tok/sNot rated
Qwen3 32B3-bit24K33 tok/sNot rated
Qwen3 4B Instruct 25078-bit128K108 tok/sNot rated
Qwen3.6 27B5-bit48K28 tok/sText and images
Qwen2.5 Coder 14B Instruct8-bit32K35 tok/sCoding
Qwen3 Coder 30B A3B Instruct5-bit16K125 tok/sCoding
Qwen3.5 27B5-bit48K28 tok/sText and images
Qwen3.5 35B A3B4-bit72K146 tok/sText and images
NVIDIA Nemotron 3 Nano 30B A3B BF165-bit176K118 tok/sNot rated
Qwen2.5 Coder 7B Instruct8-bit32K69 tok/sCoding
Mistral 7B Instruct V0.28-bit32K68 tok/sNot rated
Chandra OCR 28-bit256K87 tok/sReading documents
Qwen3 30B A3B5-bit16K125 tok/sNot rated
Qwen3 VL 2B Instruct8-bit176K188 tok/sText and images
Gemma 3 4B IT8-bit128K119 tok/sText and images
Qwen3 30B A3B Instruct 25075-bit16K125 tok/sNot rated
Llama 3.2 3B Instruct8-bit128K136 tok/sNot rated
Laguna XS.24-bit80K148 tok/sNot rated
OTel 2.0 LLM 31B IT4-bit80K28 tok/sNot rated
Qwen2.5 32B Instruct3-bit24K33 tok/sNot rated
MiniCPM V 4.7 35B A3B4-bit88K146 tok/sText and images
Qwen2.5 14B Instruct8-bit32K35 tok/sNot rated
Laguna XS 2.14-bit80K148 tok/sNot rated
Qwen3 1.7B Base8-bit32K219 tok/sNot rated
Diffusiongemma 26B A4B IT6-bit88K119 tok/sText and images
Qwen2.5 VL 32B Instruct3-bit24K32 tok/sText and images
Qwen2.5 Coder 32B Instruct3-bit24K33 tok/sCoding
Qwen2 VL 7B Instruct8-bit32K64 tok/sText and images
Ornith 1.5 9B8-bit256K57 tok/sNot rated
Granite 4.1 3B8-bit128K121 tok/sNot rated
Qwen3 VL 30B A3B Instruct4-bit44K134 tok/sText and images
Mistral 7B V0.18-bit32K68 tok/sNot rated
Meta Llama 3 8B Instruct Abliterated V38-bit8K65 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 258 are shown.

Image and video models

Image and video models that run on the Radeon RX 7900 XTX
ModelBest fitPer outputStrongest at
LTX-2.58-bit1.6 minVideo
Z-Image Turbo16-bit13 sImages
FLUX.1 [dev]16-bit26 sImages
Qwen-Image8-bit8.0 minImages
FLUX.2 [dev]5-bit4.1 minImages
Qwen-Image 2.116-bit3.5 minImages
FLUX.3 Klein16-bit19 sImages
Wan2.2 I2V A14B8-bit71 minVideo
FLUX.2 [klein] 9B16-bit6.7 sImages
Wan2.1 I2V 14B8-bit71 minVideo
Wan 2.1 T2V-14B8-bit89 minVideo
Krea 2 Turbo8-bit26 sImages
Stable Diffusion 3.5 Large16-bit1.8 minImages
Qwen Image 25128-bit8.0 minImages
Wan 2.2 TI2V-5B16-bit19 minVideo
Wan 2.2 A14B8-bit71 minVideo
HunyuanVideo 1.516-bit87 minVideo

Only just too big

These need up to a quarter more memory than the Radeon RX 7900 XTX has. Spilling into system RAM or a lower precision can get them running, slowly.

Questions

What is the best LLM for the Radeon RX 7900 XTX?

Of the open models we track, the strongest that run well on the Radeon RX 7900 XTX are Qwen3.8 27B, Muse Glimmer 30B, Qwen3.6 35B A3B. 258 text models fit it in all.

Calculated from published specs, not benchmarked. All machines