Graphics cards
Which AI models run on the RX 7900 XTX (24 GB)?
The RX 7900 XTX (24 GB) has 24 GB of memory, about 23 GB of it usable for a model, and 960 GB/s of memory bandwidth. 80 of the 110 models in our catalog run on it at 4-bit or better.
Data as of 2 November 2026
24 GBmemory
~23 GB usable for a model
960 GB/sbandwidth, sets the speed
80models run at 4-bit or better
Models that run on it
| Model | Best build | Memory | Speed |
|---|---|---|---|
| Gemma 4 31B30.7B | Q4_K_M | 21 GB | ~36 tok/s ModelFit est. |
| Qwen3-Coder-30B-A3B30.5B MoE | Q4_K_M | 20 GB | ~139 tok/s ModelFit est. |
| Nemotron 3 Nano 30B-A3B30B MoE | Q4_K_M | 19.6 GB | ~290 tok/s our est. |
| Qwen3.8-27B27.3B | Q4_K_M | 17.9 GB | ~44 tok/s ModelFit est. |
| Gemma 4 26B-A4B25.2B MoE | Q4_K_M | 17.4 GB | ~126 tok/s ModelFit est. |
| gpt-oss-20b21B MoE | MXFP4 | 15.1 GB | ~126 tok/s ModelFit est. |
| Nemotron Nano 9B v29B | FP16 | 19.4 GB | ~29 tok/s our est. |
| Qwen3.5 9B Instruct9B | Q8_0 | 12.0 GB | ~75 tok/s ModelFit est. |
| Qwen2.5 7B Instruct7B | Q4_K_M | 6.6 GB | ~116 tok/s ModelFit est. |
| Gemma 3 4B Instruct4B | Q8_0 | 5.0 GB | ~121 tok/s ModelFit est. |
| Qwen3.5 35B-A3B Instruct35B MoE | Q4_K_M | 21 GB | ~168 tok/s ModelFit est. |
| Ornith 1.0 35B35B | Q4_K_M | 23 GB | ~168 tok/s ModelFit est. |
| Nanbeige4.3 33B33.1B | Q4_K_M | 22 GB | no estimate |
| Laguna XS 2.133B | Q4_K_M | 22 GB | ~160 tok/s ModelFit est. |
| Qwen3 32B32B | Q4_K_M | 22 GB | ~35 tok/s ModelFit est. |
| DeepSeek-R1 Distill Qwen 32B32B | Q4_K_M | 21 GB | ~35 tok/s ModelFit est. |
| Olmo 3.1 32B Instruct32B | Q4_K_M | 21 GB | ~36 tok/s ModelFit est. |
| Cohere North Mini Code30B | Q4_K_M | 20 GB | no estimate |
| Nemotron 3.5 Lightning 30B-A3B30B MoE | IQ4_XS | 20 GB | ~279 tok/s our est. |
| GLM-4.7-Flash 30B-A3B30B MoE | Q4_K_M | 20 GB | ~132 tok/s ModelFit est. |
| Granite Switch 4.1 30B30B | Q4_K_M | 23 GB | no estimate |
| Muse Glimmer 30B29.8B | Q4_K_M | 18.3 GB | ~44 tok/s ModelFit est. |
| Granite 4.2 30B29.3B | Q4_K_M | 17.9 GB | ~39 tok/s ModelFit est. |
| AREX-2 27B27.4B | Q4_K_M | 18.6 GB | ~43 tok/s ModelFit est. |
| Gemma 2 27B Instruct27B | Q4_K_M | 23 GB | ~39 tok/s ModelFit est. |
| Gemma 3 27B Instruct27B | Q4_K_M | 23 GB | ~41 tok/s ModelFit est. |
| Qwen3.5 27B Instruct27B | Q4_K_M | 17.4 GB | ~44 tok/s ModelFit est. |
| Qwen3.6 27B27B | Q4_K_M | 19.4 GB | ~44 tok/s ModelFit est. |
| Jebadiah 27B27B | Q5_K_M | 21 GB | ~38 tok/s ModelFit est. |
| Mistral Small 3.124B | Q4_K_M | 16.4 GB | ~35 tok/s our est. |
| LFM2 24B-A2B Instruct24B MoE | Q4_K_M | 15.3 GB | ~176 tok/s ModelFit est. |
| Devstral Small 2 24B24B | Q4_K_M | 16.6 GB | ~45 tok/s ModelFit est. |
| Mistral Small 22B22B | Q4_K_M | 18.4 GB | ~51 tok/s ModelFit est. |
| Qwen2.5 14B Instruct14B | Q4_K_M | 12.3 GB | ~69 tok/s ModelFit est. |
| Phi-3 Medium 14B14B | Q4_K_M | 12.3 GB | ~48 tok/s our est. |
| Phi-4 14B14B | Q8_0 | 15.9 GB | ~44 tok/s ModelFit est. |
| Qwen2.5 Coder 14B14B | Q4_K_M | 12.3 GB | ~69 tok/s ModelFit est. |
| Qwen3 14B14B | Q8_0 | 17.3 GB | ~45 tok/s ModelFit est. |
| DeepSeek-R1 Distill Qwen 14B14B | Q8_0 | 16.0 GB | ~45 tok/s ModelFit est. |
| Ministral 3 14B13.5B | Q4_K_M | 10.3 GB | ~69 tok/s ModelFit est. |
| Mellum2.1 12B-A2.5B Thinking12.1B MoE | Q4_K_M | 9.3 GB | ~315 tok/s our est. |
| Mistral Nemo 12B12B | Q8_0 | 13.4 GB | ~53 tok/s ModelFit est. |
| Gemma 3 12B Instruct12B | Q4_K_M | 10.7 GB | ~79 tok/s ModelFit est. |
| Gemma 4 12B12B | Q8_0 | 14.1 GB | ~52 tok/s ModelFit est. |
| GLM-4-9B-Chat9.4B | Q8_0 | 11.2 GB | ~68 tok/s ModelFit est. |
| Gemma 2 9B Instruct9B | Q4_K_M | 8.2 GB | ~84 tok/s ModelFit est. |
| Ornith 1.0 9B9B | Q4_K_M | 6.7 GB | ~109 tok/s ModelFit est. |
| MiMo-V2.6-Distill-Qwen-9B9B | Q4_K_M | 6.9 GB | ~106 tok/s ModelFit est. |
| Jebadiah 9B9B | Q8_0 | 11.0 GB | ~73 tok/s ModelFit est. |
| Granite 4.2 8B8.8B | Q4_K_M | 6.1 GB | ~99 tok/s ModelFit est. |
| Granite Switch 4.1 8B8.8B | Q4_K_M | 7.6 GB | ~85 tok/s ModelFit est. |
| MiniCPM-V 4.5 8B8.7B | Q4_K_M | 6.8 GB | ~106 tok/s ModelFit est. |
| Ministral 3 8B8.4B | Q4_K_M | 7.1 GB | ~94 tok/s ModelFit est. |
| LFM2.5 8B-A1B8.3B MoE | Q4_K_M | 6.6 GB | ~202 tok/s ModelFit est. |
| Llama 3.1 8B Instruct8B | Q8_0 | 9.2 GB | ~74 tok/s ModelFit est. |
| Qwen3 8B8B | Q8_0 | 9.3 GB | ~72 tok/s ModelFit est. |
| Granite 4.1 8B Instruct8B | Q4_K_M | 6.6 GB | ~96 tok/s ModelFit est. |
| Mistral 7B Instruct7B | Q8_0 | 8.4 GB | ~77 tok/s ModelFit est. |
| Qwen2.5 Coder 7B7B | Q4_K_M | 6.6 GB | ~116 tok/s ModelFit est. |
| DeepSeek-R1 Distill Qwen 7B7B | Q8_0 | 8.7 GB | ~80 tok/s ModelFit est. |
| Gemma 4 E4B4.5B | Q8_0 | 8.7 GB | ~109 tok/s ModelFit est. |
| Qwen3.5 4B Instruct4B | Q8_0 | 5.4 GB | ~116 tok/s ModelFit est. |
| NVIDIA Nemotron 3.5 Bolt E4B4B | Q4_K_M | 9.1 GB | ~129 tok/s ModelFit est. |
| Phi-3 Mini 3.8B3.8B | Q4_K_M | 4.3 GB | ~165 tok/s our est. |
| Phi-4 Mini 3.8B3.8B | Q8_0 | 4.9 GB | ~116 tok/s ModelFit est. |
| Granite 4.2 3B3.7B | Q4_K_M | 3.2 GB | ~169 tok/s ModelFit est. |
| Ministral 3 3B3.4B | Q4_K_M | 4.1 GB | ~140 tok/s ModelFit est. |
| Qwen2.5 3B Instruct3B | Q4_K_M | 3.6 GB | ~182 tok/s ModelFit est. |
| Llama 3.2 3B Instruct3B | Q4_K_M | 3.6 GB | ~166 tok/s ModelFit est. |
| Granite 4.1 3B Instruct3B | Q4_K_M | 3.0 GB | ~168 tok/s ModelFit est. |
| Gemma 4 E2B2.3B | Q8_0 | 5.7 GB | ~182 tok/s ModelFit est. |
| Gemma 2 2B Instruct2B | Q4_K_M | 2.8 GB | ~179 tok/s ModelFit est. |
| Qwen3.5 2B Instruct2B | Q4_K_M | 2.8 GB | ~183 tok/s ModelFit est. |
| LFM3 1.7B1.73B | Q4_K_M | 2.1 GB | ~231 tok/s ModelFit est. |
| Qwen2.5 1.5B Instruct1.5B | Q4_K_M | 2.5 GB | ~234 tok/s ModelFit est. |
| Llama 3.2 1B Instruct1B | Q4_K_M | 2.0 GB | ~245 tok/s ModelFit est. |
| Gemma 3 1B Instruct1B | Q8_0 | 2.0 GB | ~234 tok/s ModelFit est. |
| Qwen3.5 0.8B Instruct0.8B | Q4_K_M | 1.8 GB | ~252 tok/s ModelFit est. |
| Qwen2.5 0.5B Instruct0.5B | Q4_K_M | 1.8 GB | ~289 tok/s ModelFit est. |
| SmolLM2 360M0.36B | Q4_K_M | 1.5 GB | ~249 tok/s ModelFit est. |
Speeds come from ModelFit's estimates (CC BY 4.0) where it covers the model, otherwise from our bandwidth formula. Method
Fits only at 3-bit or lower
Too big for this machine
Sage-A1.5 (23 GB), Qwen3.6-35B-A3B (24 GB), Qwen3 30B (24 GB), Flint 40B-A2.5B (24 GB), Poolside Laguna XS.2 (25 GB), NVIDIA Nemotron Cascade 2 30B-A3B (26 GB), Mixtral 8x7B Instruct (32 GB), Llama 3.3 70B (44 GB), Llama 3.1 70B Instruct (44 GB), DeepSeek-R1 Distill Llama 70B (44 GB), Kolibri 1 (46 GB), Qwen3-Next 80B-A3B (53 GB) and more.
Questions
What is the best AI model for the RX 7900 XTX (24 GB)?
Gemma 4 31B at Q4_K_M: the largest model on our shortlist of current models that runs at 4-bit or better at a usable speed (about 36 tok/s, ModelFit est.).
How much memory can a model use on the RX 7900 XTX (24 GB)?
About 23 GB of the 24 GB.
How fast is the RX 7900 XTX (24 GB) for local LLMs?
Decode speed is limited by memory bandwidth: 960 GB/s. A model that reads 10 GB per token runs at under 96 tokens per second.