Graphics cards
Which AI models run on the 2x RX 7900 XTX (48 GB)?
The 2x RX 7900 XTX (48 GB) has 48 GB of memory, about 46 GB of it usable for a model, and 960 GB/s of memory bandwidth. 92 of the 110 models in our catalog run on it at 4-bit or better.
Data as of 2 November 2026
48 GBmemory
~46 GB usable for a model
960 GB/sbandwidth, sets the speed
92models run at 4-bit or better
Two cards in one PC, model split across both. Each token still reads at one card's bandwidth.
Our pick for this machine Llama 3.3 70B The largest model on our shortlist of current models that runs at 4-bit or better and 10 tok/s or more.Models that run on it
| Model | Best build | Memory | Speed |
|---|---|---|---|
| Llama 3.3 70B70.6B | Q4_K_M | 44 GB | ~13 tok/s our est. |
| Llama 3.3 Nemotron Super 49B49B | Q6_K | 42 GB | ~13 tok/s our est. |
| Qwen3.6-35B-A3B35B MoE | Q8_0 | 41 GB | ~117 tok/s ModelFit est. |
| Gemma 4 31B30.7B | Q8_0 | 33 GB | ~22 tok/s ModelFit est. |
| Qwen3-Coder-30B-A3B30.5B MoE | Q8_0 | 34 GB | ~97 tok/s ModelFit est. |
| Nemotron 3 Nano 30B-A3B30B MoE | Q8_0 | 34 GB | ~166 tok/s our est. |
| Qwen3.8-27B27.3B | Q8_0 | 29 GB | ~27 tok/s ModelFit est. |
| Gemma 4 26B-A4B25.2B MoE | Q8_0 | 30 GB | ~92 tok/s ModelFit est. |
| gpt-oss-20b21B MoE | MXFP4 | 15.1 GB | ~122 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 | ~73 tok/s ModelFit est. |
| Qwen2.5 7B Instruct7B | Q4_K_M | 6.6 GB | ~113 tok/s ModelFit est. |
| Gemma 3 4B Instruct4B | Q8_0 | 5.0 GB | ~116 tok/s ModelFit est. |
| Kolibri 178.1B | IQ4_XS | 46 GB | no estimate |
| Llama 3.1 70B Instruct70B | Q4_K_M | 44 GB | ~13 tok/s our est. |
| DeepSeek-R1 Distill Llama 70B70B | Q4_K_M | 44 GB | ~13 tok/s our est. |
| Mixtral 8x7B Instruct46.7B MoE | Q4_K_M | 32 GB | ~54 tok/s ModelFit est. |
| Flint 40B-A2.5B39.8B MoE | NVFP4 | 24 GB | ~166 tok/s ModelFit est. |
| Sage-A1.535.95B | Q4_K_M | 23 GB | ~158 tok/s ModelFit est. |
| Qwen3.5 35B-A3B Instruct35B MoE | Q8_0 | 41 GB | ~117 tok/s ModelFit est. |
| Ornith 1.0 35B35B | Q4_K_M | 23 GB | ~160 tok/s ModelFit est. |
| Nanbeige4.3 33B33.1B | Q8_0 | 37 GB | ~22 tok/s ModelFit est. |
| Poolside Laguna XS.233B | Q4_K_M | 25 GB | no estimate |
| Laguna XS 2.133B | Q4_K_M | 22 GB | ~152 tok/s ModelFit est. |
| Qwen3 32B32B | Q8_0 | 37 GB | ~21 tok/s ModelFit est. |
| DeepSeek-R1 Distill Qwen 32B32B | Q8_0 | 37 GB | ~21 tok/s ModelFit est. |
| Olmo 3.1 32B Instruct32B | Q4_K_M | 21 GB | ~35 tok/s ModelFit est. |
| Qwen3 30B30B | Q8_0 | 32 GB | ~96 tok/s ModelFit est. |
| NVIDIA Nemotron Cascade 2 30B-A3B30B MoE | Q6_K | 26 GB | ~220 tok/s our est. |
| Cohere North Mini Code30B | Q4_K_M | 20 GB | no estimate |
| Nemotron 3.5 Lightning 30B-A3B30B MoE | Q4_K_M | 27 GB | ~117 tok/s ModelFit est. |
| GLM-4.7-Flash 30B-A3B30B MoE | Q4_K_M | 20 GB | ~127 tok/s ModelFit est. |
| Granite Switch 4.1 30B30B | Q4_K_M | 23 GB | ~33 tok/s ModelFit est. |
| Muse Glimmer 30B29.8B | Q4_K_M | 18.3 GB | ~43 tok/s ModelFit est. |
| Granite 4.2 30B29.3B | Q4_K_M | 17.9 GB | ~38 tok/s ModelFit est. |
| AREX-2 27B27.4B | Q4_K_M | 18.6 GB | ~42 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 | Q8_0 | 29 GB | ~27 tok/s ModelFit est. |
| Qwen3.6 27B27B | Q8_0 | 32 GB | ~27 tok/s ModelFit est. |
| Jebadiah 27B27B | Q8_0 | 31 GB | ~27 tok/s ModelFit est. |
| Mistral Small 3.124B | Q8_0 | 25 GB | ~28 tok/s ModelFit est. |
| LFM2 24B-A2B Instruct24B MoE | Q4_K_M | 15.3 GB | ~167 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 | ~50 tok/s ModelFit est. |
| Qwen2.5 14B Instruct14B | Q4_K_M | 12.3 GB | ~68 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 | ~68 tok/s ModelFit est. |
| Qwen3 14B14B | Q8_0 | 17.3 GB | ~44 tok/s ModelFit est. |
| DeepSeek-R1 Distill Qwen 14B14B | Q8_0 | 16.0 GB | ~44 tok/s ModelFit est. |
| Ministral 3 14B13.5B | Q4_K_M | 10.3 GB | ~68 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 | ~52 tok/s ModelFit est. |
| Gemma 3 12B Instruct12B | Q4_K_M | 10.7 GB | ~77 tok/s ModelFit est. |
| Gemma 4 12B12B | Q8_0 | 14.1 GB | ~51 tok/s ModelFit est. |
| GLM-4-9B-Chat9.4B | Q8_0 | 11.2 GB | ~67 tok/s ModelFit est. |
| Gemma 2 9B Instruct9B | Q4_K_M | 8.2 GB | ~82 tok/s ModelFit est. |
| Ornith 1.0 9B9B | Q4_K_M | 6.7 GB | ~106 tok/s ModelFit est. |
| MiMo-V2.6-Distill-Qwen-9B9B | Q4_K_M | 6.9 GB | ~103 tok/s ModelFit est. |
| Jebadiah 9B9B | Q8_0 | 11.0 GB | ~72 tok/s ModelFit est. |
| Granite 4.2 8B8.8B | Q4_K_M | 6.1 GB | ~96 tok/s ModelFit est. |
| Granite Switch 4.1 8B8.8B | Q4_K_M | 7.6 GB | ~83 tok/s ModelFit est. |
| MiniCPM-V 4.5 8B8.7B | Q4_K_M | 6.8 GB | ~103 tok/s ModelFit est. |
| Ministral 3 8B8.4B | Q4_K_M | 7.1 GB | ~92 tok/s ModelFit est. |
| LFM2.5 8B-A1B8.3B MoE | Q4_K_M | 6.6 GB | ~190 tok/s ModelFit est. |
| Llama 3.1 8B Instruct8B | Q8_0 | 9.2 GB | ~73 tok/s ModelFit est. |
| Qwen3 8B8B | Q8_0 | 9.3 GB | ~71 tok/s ModelFit est. |
| Granite 4.1 8B Instruct8B | Q4_K_M | 6.6 GB | ~94 tok/s ModelFit est. |
| Mistral 7B Instruct7B | Q8_0 | 8.4 GB | ~76 tok/s ModelFit est. |
| Qwen2.5 Coder 7B7B | Q4_K_M | 6.6 GB | ~113 tok/s ModelFit est. |
| DeepSeek-R1 Distill Qwen 7B7B | Q8_0 | 8.7 GB | ~78 tok/s ModelFit est. |
| Gemma 4 E4B4.5B | Q8_0 | 8.7 GB | ~106 tok/s ModelFit est. |
| Qwen3.5 4B Instruct4B | Q8_0 | 5.4 GB | ~112 tok/s ModelFit est. |
| NVIDIA Nemotron 3.5 Bolt E4B4B | Q4_K_M | 9.1 GB | ~124 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 | ~112 tok/s ModelFit est. |
| Granite 4.2 3B3.7B | Q4_K_M | 3.2 GB | ~161 tok/s ModelFit est. |
| Ministral 3 3B3.4B | Q4_K_M | 4.1 GB | ~135 tok/s ModelFit est. |
| Qwen2.5 3B Instruct3B | Q4_K_M | 3.6 GB | ~173 tok/s ModelFit est. |
| Llama 3.2 3B Instruct3B | Q4_K_M | 3.6 GB | ~159 tok/s ModelFit est. |
| Granite 4.1 3B Instruct3B | Q4_K_M | 3.0 GB | ~160 tok/s ModelFit est. |
| Gemma 4 E2B2.3B | Q8_0 | 5.7 GB | ~173 tok/s ModelFit est. |
| Gemma 2 2B Instruct2B | Q4_K_M | 2.8 GB | ~170 tok/s ModelFit est. |
| Qwen3.5 2B Instruct2B | Q4_K_M | 2.8 GB | ~174 tok/s ModelFit est. |
| LFM3 1.7B1.73B | Q4_K_M | 2.1 GB | ~216 tok/s ModelFit est. |
| Qwen2.5 1.5B Instruct1.5B | Q4_K_M | 2.5 GB | ~219 tok/s ModelFit est. |
| Llama 3.2 1B Instruct1B | Q4_K_M | 2.0 GB | ~228 tok/s ModelFit est. |
| Gemma 3 1B Instruct1B | Q8_0 | 2.0 GB | ~219 tok/s ModelFit est. |
| Qwen3.5 0.8B Instruct0.8B | Q4_K_M | 1.8 GB | ~234 tok/s ModelFit est. |
| Qwen2.5 0.5B Instruct0.5B | Q4_K_M | 1.8 GB | ~266 tok/s ModelFit est. |
| SmolLM2 360M0.36B | Q4_K_M | 1.5 GB | ~232 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
Too big for this machine
Qwen3-Next 80B-A3B (53 GB), Nemotron 3 Super 120B-A12B (67 GB), gpt-oss-120b (68 GB), Llama 4 Scout (70 GB), Ling-3.0-flash (71 GB), Qwen3.5 122B-A10B Instruct (75 GB), Nemotron 3.5 Super VL 120B-A12B (76 GB), Devstral 2 123B (78 GB), Ornith 1.5 122B (79 GB), Qwen3.8-Flash-Next (93 GB), Laguna S 2.1 (99 GB), Qwen3-235B-A22B (134 GB) and more.
Questions
What is the best AI model for the 2x RX 7900 XTX (48 GB)?
Llama 3.3 70B 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 13 tok/s, our est.).
How much memory can a model use on the 2x RX 7900 XTX (48 GB)?
About 46 GB of the 48 GB. Two cards in one PC, model split across both. Each token still reads at one card's bandwidth.
How fast is the 2x RX 7900 XTX (48 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.