AMD Radeon RX 9070
16 GB · 640 GB/s · 220 W
In short
The AMD Radeon RX 9070 has 16 GB of GDDR6 memory with 640 GB/s of bandwidth, which sets how fast it generates text. At 128k context it runs 21 of the 125 open models we track fully in memory; the most capable is Gemma 4 12B at Q6_K, at about 29 tok/s (estimated). We have no current price for the AMD Radeon RX 9070 in the US.
Updated . Speeds are estimates; how they are worked out.
Prices
Every retailer we track in the US, cheapest first. Buy links may earn us a commission.
No retailer in this market lists it right now.
What it runs
Every model in the catalogue, with the best quant for this hardware. Estimates, not measurements.
| Model | Quant | Fits | Memory | Max context | Decode @128k | Prefill | Good for |
|---|---|---|---|---|---|---|---|
| Gemma 4 12B | Q6_K | fits | 13.0 GB | 231k256k with q8 cache | 29tok/sest. | 381 tok/s | chat, coding, long documents |
| Qwen 3.5 9B Alibaba Qwen | Q8_0 | fits | 14.5 GB | 147k243k with q8 cache | 25tok/sest. | 453 tok/s | chat, coding, long documents |
| Qwen 3.5 4B Alibaba Qwen | Q8_0 | fits | 9.8 GB | 256k | 40tok/sest. | 706 tok/s | chat, coding, long documents |
| Gemma 4 E4B | Q8_0 | fits | 12.7 GB | 128k | 29tok/sest. | 635 tok/s | chat, coding, long documents |
| LFM2.5 8B-A1B Liquid AI | UD-Q6_K | fits | 14.2 GB | 125k | 29tok/sest. | 849 tok/s | chat, coding, long documents |
| Qwen2.5 0.5B Alibaba Qwen | Q8_0 | fits | 3.5 GB | 32k | 164tok/sest. | 1907 tok/s | chat, coding, long documents |
| Qwen3.5 0.8B Alibaba Qwen | Q8_0 | fits | 3.8 GB | 256k | 144tok/sest. | 2543 tok/s | chat, coding, agents, long documents |
| Qwen3.5 2B Alibaba Qwen | Q5_K_M | fits | 4.4 GB | 256k | 115tok/sest. | 1626 tok/s | chat, coding, agents, long documents |
| DeepSeek R1 Distill Qwen 1.5B DeepSeek | Q8_0 | fits | 6.8 GB | 128k | 62tok/sest. | 852 tok/s | chat, coding, long documents |
| Mellum2.1 12B A2.5B Thinking JetBrains | Q6_K | fits | 14.0 GB | 128k | 49tok/sest. | 982 tok/s | chat, coding, agents, long documents |
| Spark X2.5 4B XHToken | mlx-4bit | fits | 8.9 GB | 280k470k with q8 cache | 44tok/sest. | 698 tok/s | chat, coding, agents, long documents |
| MiniCPM5 2B openbmb | Q8_0 | fits | 9.3 GB | 128k | 42tok/sest. | 458 tok/s | chat, coding, agents, long documents |
| FrogNano 4B 2609 Microsoft | Q8_0 | fits | 9.9 GB | 256k | 39tok/sest. | 706 tok/s | chat, coding, agents, long documents |
| Kimi VL A3B Thinking 2506 Moonshot AI | Q4_K_M | fits | 15.2 GB | 128k | 35tok/sest. | 617 tok/s | chat, coding, long documents |
| LFM2.5 2.6B Liquid AI | Q8_0 | fits | 11.7 GB | 128k | 32tok/sest. | 588 tok/s | chat, coding, agents, long documents |
| LFM2.5 VL 3B Liquid AI | Q8_0 | fits | 11.7 GB | 32k | 32tok/sest. | 565 tok/s | chat, coding, agents, long documents |
| GLM 4.6V Flash Z.ai | Q4_K_M | fits | 12.3 GB | 128k | 31tok/sest. | 199 tok/s | chat, coding |
| GLM 4 9B 0414 Z.ai | mlx-5bit | fits | 14.2 GB | 32k | 26tok/sest. | 205 tok/s | chat, coding |
| DeepSeek R1 Distill Qwen 7B DeepSeek | Q6_K | fits | 14.4 GB | 128k | 26tok/sest. | 304 tok/s | chat, coding, long documents |
| granite vision 4.1 4b IBM | Q8_0 | fits | 14.9 GB | 128k | 25tok/sest. | 363 tok/s | chat, coding, long documents |
| granite 4.2 3b IBM | Q8_0 | fits | 15.2 GB | 128k | 24tok/sest. | 370 tok/s | chat, coding, agents, long documents |
| Qwen 3.8 27B Alibaba Qwen | UD-Q4_K_M | offload | 25.0 GB | – | 4tok/sest. | 116 tok/s | – |
| Qwen 3.6 27B Alibaba Qwen | Q4_K_M | offload | 25.4 GB | – | 4tok/sest. | 116 tok/s | – |
| Gemma 4 31B | Q4_K_M | offload | 29.4 GB | – | 3tok/sest. | 102 tok/s | – |
| Qwen 3.6 35B-A3B Alibaba Qwen | MXFP4 | offload | 24.3 GB | – | 14tok/sest. | 573 tok/s | chat, coding |
| Qwen 3.5 27B Alibaba Qwen | Q4_K_M | offload | 25.3 GB | – | 4tok/sest. | 116 tok/s | – |
| Gemma 4 26B-A4B | MXFP4 | offload | 19.7 GB | – | 16tok/sest. | 781 tok/s | chat, coding, agents, long documents |
| Qwen 3.5 35B-A3B Alibaba Qwen | Q4_K_M | offload | 24.6 GB | – | 14tok/sest. | 570 tok/s | chat, coding |
| GLM 4.7 Flash Z.ai | Q4_K_M | offload | 25.2 GB | – | 6tok/sest. | 266 tok/s | chat |
| Devstral Small 2 24B Mistral AI | Q4_K_M | offload | 34.9 GB | 6k11k with q8 cache | 2tok/sest. | 90 tok/s | chat |
| Nemotron 3.5 Lightning 30B-A3B NVIDIA | MXFP4 | offload | 24.0 GB | – | 22tok/sest. | 777 tok/s | chat, coding, long documents |
| gpt-oss 20B OpenAI | Q4_K_M | offload | 15.4 GB | 121k128k with q8 cache | 36tok/sest. | 589 tok/s | chat, coding, agents, long documents |
| Kimi Linear 48B-A3B Moonshot AI | Q4_K_M | offload | 30.6 GB | – | 18tok/sest. | 386 tok/s | chat, coding, long documents |
| Qwen3 Coder 30B-A3B Alibaba Qwen | Q4_K_M | offload | 30.8 GB | – | 4tok/sest. | 149 tok/s | chat |
| Gemma 3 27B | Q4_K_M | offload | 27.4 GB | – | 4tok/sest. | 136 tok/s | – |
| Mistral Small 3.2 24B Mistral AI | Q4_K_M | offload | 34.9 GB | 6k11k with q8 cache | 2tok/sest. | 90 tok/s | chat |
| Ministral 3 14B Mistral AI | Q4_K_M | offload | 29.2 GB | 41k75k with q8 cache | 3tok/sest. | 124 tok/s | chat, coding |
| Qwen3 14B Alibaba Qwen | Q4_K_M | offload | 29.9 GB | 37k40k with q8 cache | 3tok/sest. | 105 tok/s | chat, coding |
| DeepSeek R1 0528 Qwen3 8B DeepSeek | Q4_K_M | offload | 24.2 GB | 66k120k with q8 cache | 5tok/sest. | 175 tok/s | chat, coding |
| Qwen3 8B Alibaba Qwen | Q4_K_M | offload | 24.2 GB | 40k | 5tok/sest. | 175 tok/s | chat, coding |
| Phi-4 14B Microsoft | Q4_K_M | offload | 34.8 GB | 16k | 2tok/sest. | 98 tok/s | chat, coding |
| Ministral 3 8B Mistral AI | Q4_K_M | offload | 23.4 GB | 68k125k with q8 cache | 5tok/sest. | 180 tok/s | chat, coding |
| Granite 4.2 8B IBM | Q4_K_M | offload | 26.7 GB | 57k103k with q8 cache | 4tok/sest. | 151 tok/s | chat, coding |
| Llama 3.1 8B Meta | Q4_K_M | offload | 22.1 GB | 75k128k with q8 cache | 6tok/sest. | 200 tok/s | chat, coding |
| Kimi VL A3B Moonshot AI | mlx-5bit | offload | 16.4 GB | 95k128k with q8 cache | 24tok/sest. | 587 tok/s | chat, coding, agents, long documents |
| Ministral 3 3B Instruct 2512 Mistral AI | Q4_K_M | offload | 16.5 GB | 115k207k with q8 cache | 15tok/sest. | 336 tok/s | chat, coding, agents |
| GLM 4.1V 9B Thinking Z.ai | mlx-8bit | offload | 17.5 GB | 64k | 11tok/sest. | 182 tok/s | chat, coding |
| Muse Glimmer 30B meta-models | Q4_K_M | offload | 19.4 GB | – | 8tok/sest. | 145 tok/s | – |
| DeepSeek R1 Distill Llama 8B DeepSeek | Q4_K_M | offload | 22.1 GB | 75k128k with q8 cache | 6tok/sest. | 200 tok/s | chat, coding |
| LFM2.5 24B A2B Liquid AI | Q4_K_M | offload | 25.8 GB | 1k2k with q8 cache | 5tok/sest. | 366 tok/s | chat, coding |
| Ministral 3 8B Reasoning 2512 Mistral AI | Q4_K_M | offload | 23.4 GB | 68k125k with q8 cache | 5tok/sest. | 180 tok/s | chat, coding |
| Qwen3.8 27B OBLITERATED OBLITERATUS | mlx-4bit | offload | 28.9 GB | – | 3tok/sest. | 109 tok/s | – |
| Ministral 3 14B Reasoning 2512 Mistral AI | Q4_K_M | offload | 29.2 GB | 41k75k with q8 cache | 3tok/sest. | 124 tok/s | chat, coding |
| DeepSeek R1 Distill Qwen 14B DeepSeek | Q4_K_M | offload | 33.9 GB | 31k57k with q8 cache | 2tok/sest. | 88 tok/s | chat, coding |
| Devstral Small 2507 Mistral AI | Q4_K_M | offload | 34.9 GB | 6k11k with q8 cache | 2tok/sest. | 91 tok/s | chat |
| Magistral Small 2506 Mistral AI | Q4_K_M | offload | 34.9 GB | 6k11k with q8 cache | 2tok/sest. | 91 tok/s | chat |
| Magistral Small 2507 Mistral AI | Q4_K_M | offload | 34.9 GB | 6k11k with q8 cache | 2tok/sest. | 91 tok/s | chat |
| Magistral Small 2509 Mistral AI | Q4_K_M | offload | 34.9 GB | 6k11k with q8 cache | 2tok/sest. | 90 tok/s | chat |
| DeepSeek V4 Flash DeepSeek | UD-Q2_K_XL | won't run | 102.5 GB | – | – | – | – |
| GLM 5.3 Flash Z.ai | UD-Q2_K_XL | won't run | 104.3 GB | – | – | – | – |
| Qwen 3.8 Flash Next Alibaba Qwen | UD-Q2_K_XL | won't run | 78.1 GB | – | – | – | – |
| MiniMax M2.7 MiniMax | UD-Q2_K_XL | won't run | 102.7 GB | – | – | – | – |
| Qwen 3.5 122B-A10B Alibaba Qwen | UD-Q2_K_XL | won't run | 43.7 GB | – | – | – | – |
| Nemotron 3 Super 120B-A12B NVIDIA | UD-Q2_K_XL | won't run | 53.6 GB | – | – | – | – |
| Qwen3 Coder Next Alibaba Qwen | Q4_K_M | won't run | 49.8 GB | – | – | – | – |
| gpt-oss 120B OpenAI | Q4_K_M | won't run | 64.5 GB | – | – | – | – |
| Mistral Small 4 119B Mistral AI | UD-Q2_K_XL | won't run | 41.8 GB | – | – | – | – |
| GLM 4.5 Air Z.ai | Q2_K | won't run | 66.8 GB | – | – | – | – |
| Qwen3 Next 80B-A3B Alibaba Qwen | Q4_K_M | won't run | 49.8 GB | – | – | – | – |
| Qwen3 32B Alibaba Qwen | Q4_K_M | won't run | 52.0 GB | – | – | – | – |
| DeepSeek R1 Distill Qwen 32B DeepSeek | Q4_K_M | won't run | 52.0 GB | – | – | – | – |
| Llama 3.3 70B Meta | Q4_K_M | won't run | 81.2 GB | – | – | – | – |
| DeepSeek R1 DeepSeek | Q4_K_M | won't run | 386.8 GB | – | – | – | – |
| DeepSeek R1 0528 DeepSeek | Q4_K_M | won't run | 387.3 GB | – | – | – | – |
| DeepSeek R1 Distill Llama 70B DeepSeek | Q4_K_M | won't run | 81.2 GB | – | – | – | – |
| DeepSeek R1 Zero DeepSeek | Q4_K_M | won't run | 386.8 GB | – | – | – | – |
| DeepSeek V3 DeepSeek | Q4_K_M | won't run | 386.8 GB | – | – | – | – |
| DeepSeek V3 0324 DeepSeek | Q4_K_M | won't run | 386.8 GB | – | – | – | – |
| DeepSeek V3.1 DeepSeek | Q4_K_M | won't run | 387.7 GB | – | – | – | – |
| DeepSeek V3.1 Terminus DeepSeek | Q4_K_M | won't run | 387.7 GB | – | – | – | – |
| DeepSeek V3.2 DeepSeek | Q4_K_M | won't run | 387.7 GB | – | – | – | – |
| DeepSeek V4 Flash 0731 DeepSeek | MXFP4 | won't run | 157.9 GB | – | – | – | – |
| DeepSeek V4 Flash Vision Exp DeepSeek | MXFP4 | won't run | 166.7 GB | – | – | – | – |
| Devstral 2 123B Instruct 2512 Mistral AI | Q4_K_M | won't run | 115.3 GB | – | – | – | – |
| GLM 4.5 Z.ai | Q4_K_M | won't run | 249.1 GB | – | – | – | – |
| GLM 4.5V Z.ai | Q4_K_M | won't run | 83.8 GB | – | – | – | – |
| GLM 4.6 Z.ai | Q4_K_M | won't run | 248.3 GB | – | – | – | – |
| GLM 4.6V Z.ai | Q4_K_M | won't run | 90.2 GB | – | – | – | – |
| GLM 4.7 Z.ai | Q4_K_M | won't run | 249.1 GB | – | – | – | – |
| GLM 5 Z.ai | MXFP4 | won't run | 394.9 GB | – | – | – | – |
| GLM 5.1 Z.ai | MXFP4 | won't run | 432.5 GB | – | – | – | – |
| GLM 5.2 Z.ai | UD-Q4_K_M | won't run | 446.4 GB | – | – | – | – |
| GLM 5.3 Z.ai | Q8_0 | won't run | 758.8 GB | – | – | – | – |
| Kimi Dev 72B Moonshot AI | Q6_K | won't run | 101.5 GB | – | – | – | – |
| Kimi K2 Moonshot AI | Q4_K_M | won't run | 588.3 GB | – | – | – | – |
| Kimi K2 Instruct 0905 Moonshot AI | Q4_K_M | won't run | 588.7 GB | – | – | – | – |
| Kimi K2 Thinking Moonshot AI | Q4_K_M | won't run | 588.7 GB | – | – | – | – |
| Kimi K2.5 Moonshot AI | Q4_K_M | won't run | 588.7 GB | – | – | – | – |
| MiniMax M2 MiniMax | Q4_K_M | won't run | 161.4 GB | – | – | – | – |
| MiniMax M2.1 MiniMax | Q4_K_M | won't run | 161.4 GB | – | – | – | – |
| MiniMax M2.5 MiniMax | MXFP4 | won't run | 147.8 GB | – | – | – | – |
| MiniMax M3 MiniMax | MXFP4 | won't run | 256.2 GB | – | – | – | – |
| Mistral Medium 3.5 128B Mistral AI | Q4_K_M | won't run | 115.3 GB | – | – | – | – |
| granite 4.2 30b IBM | Q4_K_M | won't run | 50.3 GB | – | – | – | – |
Specifications
From the manufacturer's published specifications; sources are linked where we have them.
- GPU
- Radeon RX 9070
- Memory
- 16 GB GDDR6
- Memory bus
- 256-bit
- Memory bandwidth
- 640 GB/s
- FP16 compute
- 36.1 TFLOPS
- Software
- ROCm
- Board power
- 220 W
- PCIe generation
- Gen 5
- PCIe lanes
- x16
Compare with
Questions
- How much memory does the AMD Radeon RX 9070 have for AI models?
- The AMD Radeon RX 9070 has 16 GB of GDDR6 memory with 640 GB/s of bandwidth, which sets how fast it generates text.
- What is the best model the AMD Radeon RX 9070 can run locally?
- Gemma 4 12B (Q6_K) fits fully in its memory at 128k context and generates at about 29 tok/s, an estimate from its memory bandwidth. 21 of the 125 models we track fit at that context.
- Can the AMD Radeon RX 9070 run DeepSeek V4 Flash?
- No: even at UD-Q2_K_XL, DeepSeek V4 Flash needs about 102 GB at 128k context, more than the AMD Radeon RX 9070 and system RAM can hold.
- How much does the AMD Radeon RX 9070 cost in the US?
- We have no current price for the AMD Radeon RX 9070 in the US.
Speeds on this page are for 128k context. See the best hardware for local AI at every budget.