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 8k context it runs 61 of the 125 open models we track fully in memory; the most capable is Qwen 3.8 27B at Q8_0, at about 12 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 two of them run
Every model in the catalogue, with the best quant for this hardware. Estimates, not measurements.
| Model | Quant | Fits | Memory | Max context | Decode @8k | Prefill | Good for |
|---|---|---|---|---|---|---|---|
| Qwen 3.8 27B Alibaba Qwen | Q8_0 | fits | 29.0 GB | 27k48k with q8 cache | 12tok/sest. | 221 tok/s | – |
| Qwen 3.6 27B Alibaba Qwen | Q8_0 | fits | 28.6 GB | 34k59k with q8 cache | 12tok/sest. | 221 tok/s | – |
| Gemma 4 31B | Q6_K | fits | 26.2 GB | 57k110k with q8 cache | 13tok/sest. | 193 tok/s | – |
| Qwen 3.6 35B-A3B Alibaba Qwen | Q6_K | fits | 30.4 GB | 8k13k with q8 cache | 75tok/sest. | 1894 tok/s | chat, coding, agents, long documents |
| Qwen 3.5 27B Alibaba Qwen | Q8_0 | fits | 28.6 GB | 34k59k with q8 cache | 12tok/sest. | 221 tok/s | – |
| Gemma 4 26B-A4B | Q8_0 | fits | 26.7 GB | 153k245k with q8 cache | 47tok/sest. | 1443 tok/s | chat, coding, agents, long documents |
| Qwen 3.5 35B-A3B Alibaba Qwen | Q6_K | fits | 28.4 GB | 85k133k with q8 cache | 80tok/sest. | 1894 tok/s | chat, coding, agents, long documents |
| GLM 4.7 Flash Z.ai | Q6_K | fits | 24.8 GB | 106k183k with q8 cache | 47tok/sest. | 1060 tok/s | chat, coding, agents, long documents |
| Devstral Small 2 24B Mistral AI | Q8_0 | fits | 25.9 GB | 35k64k with q8 cache | 13tok/sest. | 249 tok/s | – |
| Nemotron 3.5 Lightning 30B-A3B NVIDIA | UD-Q5_K_M | fits | 29.6 GB | 78k101k with q8 cache | 71tok/sest. | 1973 tok/s | chat, coding, agents, long documents |
| Kimi Linear 48B-A3B Moonshot AI | Q4_K_M | fits | 29.4 GB | 78k107k with q8 cache | 109tok/sest. | 1800 tok/s | chat, coding, long documents |
| gpt-oss 20B OpenAI | MXFP4 | fits | 12.8 GB | 128k | 80tok/sest. | 1371 tok/s | chat, coding, agents, long documents |
| Qwen3 Coder 30B-A3B Alibaba Qwen | Q6_K | fits | 25.5 GB | 57k102k with q8 cache | 65tok/sest. | 1370 tok/s | chat, coding, agents |
| Qwen3 32B Alibaba Qwen | Q6_K | fits | 28.4 GB | 15k29k with q8 cache | 12tok/sest. | 171 tok/s | – |
| Gemma 4 12B | Q8_0 | fits | 13.6 GB | 256k | 27tok/sest. | 503 tok/s | chat, coding, long documents |
| Qwen 3.5 9B Alibaba Qwen | Q8_0 | fits | 10.5 GB | 256k | 36tok/sest. | 637 tok/s | chat, coding, long documents |
| Mistral Small 3.2 24B Mistral AI | Q8_0 | fits | 25.9 GB | 35k64k with q8 cache | 13tok/sest. | 249 tok/s | – |
| DeepSeek R1 Distill Qwen 32B DeepSeek | Q6_K | fits | 28.4 GB | 15k29k with q8 cache | 12tok/sest. | 178 tok/s | – |
| Gemma 3 27B | Q8_0 | fits | 29.1 GB | 23k45k with q8 cache | 12tok/sest. | 224 tok/s | – |
| Ministral 3 14B Mistral AI | Q8_0 | fits | 16.0 GB | 97k176k with q8 cache | 22tok/sest. | 413 tok/s | chat |
| Qwen3 14B Alibaba Qwen | Q8_0 | fits | 17.2 GB | 40k | 21tok/sest. | 384 tok/s | chat |
| DeepSeek R1 0528 Qwen3 8B DeepSeek | Q8_0 | fits | 10.6 GB | 128k | 36tok/sest. | 672 tok/s | chat, coding |
| Qwen3 8B Alibaba Qwen | Q8_0 | fits | 10.6 GB | 40k | 36tok/sest. | 672 tok/s | chat, coding |
| Phi-4 14B Microsoft | Q8_0 | fits | 17.4 GB | 16k | 20tok/sest. | 387 tok/s | chat |
| Qwen 3.5 4B Alibaba Qwen | Q8_0 | fits | 5.8 GB | 256k | 73tok/sest. | 1282 tok/s | chat, coding, long documents |
| Ministral 3 8B Mistral AI | Q8_0 | fits | 10.8 GB | 149k256k with q8 cache | 35tok/sest. | 628 tok/s | chat, coding |
| Granite 4.2 8B IBM | Q8_0 | fits | 11.3 GB | 125k128k with q8 cache | 33tok/sest. | 623 tok/s | chat, coding |
| Gemma 4 E4B | Q8_0 | fits | 9.2 GB | 128k | 42tok/sest. | 772 tok/s | chat, coding, long documents |
| LFM2.5 8B-A1B Liquid AI | Q8_0 | fits | 10.1 GB | 125k | 144tok/sest. | 4504 tok/s | chat, coding, long documents |
| Llama 3.1 8B Meta | Q6_K | fits | 8.5 GB | 128k | 46tok/sest. | 694 tok/s | chat, coding |
| Qwen2.5 0.5B Alibaba Qwen | Q8_0 | fits | 1.9 GB | 32k | 557tok/sest. | 9426 tok/s | chat, coding, long documents |
| Qwen3.5 0.8B Alibaba Qwen | Q8_0 | fits | 2.2 GB | 256k | 378tok/sest. | 6486 tok/s | chat, coding, agents, long documents |
| Qwen3.5 2B Alibaba Qwen | Q5_K_M | fits | 2.8 GB | 256k | 226tok/sest. | 2660 tok/s | chat, coding, agents, long documents |
| DeepSeek R1 Distill Qwen 1.5B DeepSeek | Q8_0 | fits | 3.3 GB | 128k | 165tok/sest. | 2967 tok/s | chat, coding, long documents |
| MiniCPM5 2B openbmb | Q8_0 | fits | 4.2 GB | 128k | 116tok/sest. | 1961 tok/s | chat, coding, agents, long documents |
| LFM2.5 2.6B Liquid AI | Q8_0 | fits | 4.5 GB | 128k | 104tok/sest. | 1974 tok/s | chat, coding, agents, long documents |
| LFM2.5 VL 3B Liquid AI | Q8_0 | fits | 4.5 GB | 32k | 104tok/sest. | 1742 tok/s | chat, coding, agents, long documents |
| Spark X2.5 4B XHToken | mlx-4bit | fits | 4.5 GB | 639k1024k with q8 cache | 104tok/sest. | 1395 tok/s | chat, coding, agents, long documents |
| granite vision 4.1 4b IBM | Q8_0 | fits | 5.3 GB | 128k | 82tok/sest. | 1306 tok/s | chat, coding, long documents |
| Ministral 3 3B Instruct 2512 Mistral AI | Q8_0 | fits | 5.6 GB | 239k256k with q8 cache | 78tok/sest. | 1338 tok/s | chat, coding, agents, long documents |
| granite 4.2 3b IBM | Q8_0 | fits | 5.6 GB | 128k | 77tok/sest. | 1404 tok/s | chat, coding, agents, long documents |
| LFM2.5 24B A2B Liquid AI | Q8_0 | fits | 25.7 GB | 63k113k with q8 cache | 76tok/sest. | 2365 tok/s | chat, coding, agents |
| Mellum2.1 12B A2.5B Thinking JetBrains | Q8_0 | fits | 13.5 GB | 128k | 75tok/sest. | 2170 tok/s | chat, coding, agents, long documents |
| FrogNano 4B 2609 Microsoft | Q8_0 | fits | 5.9 GB | 256k | 71tok/sest. | 1282 tok/s | chat, coding, agents, long documents |
| Kimi VL A3B Thinking 2506 Moonshot AI | Q8_0 | fits | 17.4 GB | 128k | 63tok/sest. | 1830 tok/s | chat, coding, long documents |
| Kimi VL A3B Moonshot AI | mlx-8bit | fits | 18.2 GB | 128k | 60tok/sest. | 1830 tok/s | chat, coding, agents, long documents |
| DeepSeek R1 Distill Qwen 7B DeepSeek | Q8_0 | fits | 9.3 GB | 128k | 41tok/sest. | 749 tok/s | chat, coding, long documents |
| GLM 4 9B 0414 Z.ai | mlx-5bit | fits | 9.3 GB | 32k | 41tok/sest. | 588 tok/s | chat, coding |
| DeepSeek R1 Distill Llama 8B DeepSeek | Q8_0 | fits | 10.3 GB | 128k | 37tok/sest. | 694 tok/s | chat, coding |
| Ministral 3 8B Reasoning 2512 Mistral AI | Q8_0 | fits | 10.8 GB | 149k256k with q8 cache | 35tok/sest. | 628 tok/s | chat, coding |
| GLM 4.6V Flash Z.ai | Q8_0 | fits | 11.0 GB | 128k | 34tok/sest. | 543 tok/s | chat, coding |
| GLM 4.1V 9B Thinking Z.ai | mlx-8bit | fits | 12.6 GB | 64k | 29tok/sest. | 543 tok/s | chat, coding |
| Ministral 3 14B Reasoning 2512 Mistral AI | Q8_0 | fits | 16.0 GB | 97k176k with q8 cache | 22tok/sest. | 413 tok/s | chat |
| DeepSeek R1 Distill Qwen 14B DeepSeek | Q8_0 | fits | 17.5 GB | 74k128k with q8 cache | 20tok/sest. | 376 tok/s | chat |
| Qwen3.8 27B OBLITERATED OBLITERATUS | mlx-4bit | fits | 21.2 GB | 142k248k with q8 cache | 17tok/sest. | 221 tok/s | chat |
| Muse Glimmer 30B meta-models | Q6_K | fits | 23.3 GB | 128k | 15tok/sest. | 205 tok/s | – |
| Devstral Small 2507 Mistral AI | Q8_0 | fits | 25.9 GB | 35k64k with q8 cache | 13tok/sest. | 254 tok/s | – |
| Magistral Small 2506 Mistral AI | Q8_0 | fits | 25.9 GB | 35k40k with q8 cache | 13tok/sest. | 254 tok/s | – |
| Magistral Small 2507 Mistral AI | Q8_0 | fits | 25.9 GB | 35k40k with q8 cache | 13tok/sest. | 254 tok/s | – |
| Magistral Small 2509 Mistral AI | Q8_0 | fits | 25.9 GB | 35k64k with q8 cache | 13tok/sest. | 249 tok/s | – |
| granite 4.2 30b IBM | Q6_K | fits | 26.2 GB | 24k45k with q8 cache | 13tok/sest. | 201 tok/s | – |
| Qwen 3.5 122B-A10B Alibaba Qwen | UD-Q2_K_XL | offload | 40.6 GB | – | 23tok/sest. | 493 tok/s | chat |
| Qwen3 Coder Next Alibaba Qwen | Q4_K_M | offload | 46.8 GB | – | 27tok/sest. | 1157 tok/s | chat, coding |
| Nemotron 3 Super 120B-A12B NVIDIA | UD-Q2_K_XL | offload | 52.5 GB | – | 11tok/sest. | 382 tok/s | – |
| Mistral Small 4 119B Mistral AI | UD-Q2_K_XL | offload | 39.0 GB | – | 36tok/sest. | 660 tok/s | chat, coding |
| GLM 4.5 Air Z.ai | Q2_K | offload | 45.0 GB | – | 10tok/sest. | 279 tok/s | – |
| Qwen3 Next 80B-A3B Alibaba Qwen | Q4_K_M | offload | 46.8 GB | – | 34tok/sest. | 1455 tok/s | chat, coding |
| Llama 3.3 70B Meta | Q4_K_M | offload | 43.4 GB | – | 3tok/sest. | 68 tok/s | – |
| DeepSeek R1 Distill Llama 70B DeepSeek | Q4_K_M | offload | 43.4 GB | – | 3tok/sest. | 68 tok/s | – |
| DeepSeek V4 Flash DeepSeek | UD-Q2_K_XL | won't run | 92.2 GB | – | – | – | – |
| GLM 5.3 Flash Z.ai | UD-Q2_K_XL | won't run | 102.8 GB | – | – | – | – |
| Qwen 3.8 Flash Next Alibaba Qwen | UD-Q2_K_XL | won't run | 75.1 GB | – | – | – | – |
| MiniMax M2.7 MiniMax | UD-Q2_K_XL | won't run | 73.4 GB | – | – | – | – |
| gpt-oss 120B OpenAI | Q4_K_M | won't run | 60.1 GB | – | – | – | – |
| DeepSeek R1 DeepSeek | Q4_K_M | won't run | 378.5 GB | – | – | – | – |
| DeepSeek R1 0528 DeepSeek | Q4_K_M | won't run | 379.0 GB | – | – | – | – |
| DeepSeek R1 Zero DeepSeek | Q4_K_M | won't run | 378.5 GB | – | – | – | – |
| DeepSeek V3 DeepSeek | Q4_K_M | won't run | 378.5 GB | – | – | – | – |
| DeepSeek V3 0324 DeepSeek | Q4_K_M | won't run | 378.5 GB | – | – | – | – |
| DeepSeek V3.1 DeepSeek | Q4_K_M | won't run | 379.4 GB | – | – | – | – |
| DeepSeek V3.1 Terminus DeepSeek | Q4_K_M | won't run | 379.4 GB | – | – | – | – |
| DeepSeek V3.2 DeepSeek | Q4_K_M | won't run | 379.4 GB | – | – | – | – |
| DeepSeek V4 Flash 0731 DeepSeek | MXFP4 | won't run | 147.7 GB | – | – | – | – |
| DeepSeek V4 Flash Vision Exp DeepSeek | MXFP4 | won't run | 156.4 GB | – | – | – | – |
| Devstral 2 123B Instruct 2512 Mistral AI | Q4_K_M | won't run | 73.9 GB | – | – | – | – |
| GLM 4.5 Z.ai | Q4_K_M | won't run | 205.8 GB | – | – | – | – |
| GLM 4.5V Z.ai | Q4_K_M | won't run | 62.0 GB | – | – | – | – |
| GLM 4.6 Z.ai | Q4_K_M | won't run | 205.0 GB | – | – | – | – |
| GLM 4.6V Z.ai | Q4_K_M | won't run | 68.4 GB | – | – | – | – |
| GLM 4.7 Z.ai | Q4_K_M | won't run | 205.8 GB | – | – | – | – |
| GLM 5 Z.ai | MXFP4 | won't run | 384.4 GB | – | – | – | – |
| GLM 5.1 Z.ai | MXFP4 | won't run | 422.0 GB | – | – | – | – |
| GLM 5.2 Z.ai | UD-Q4_K_M | won't run | 435.9 GB | – | – | – | – |
| GLM 5.3 Z.ai | Q8_0 | won't run | 748.4 GB | – | – | – | – |
| Kimi Dev 72B Moonshot AI | Q6_K | won't run | 63.8 GB | – | – | – | – |
| Kimi K2 Moonshot AI | Q4_K_M | won't run | 580.0 GB | – | – | – | – |
| Kimi K2 Instruct 0905 Moonshot AI | Q4_K_M | won't run | 580.5 GB | – | – | – | – |
| Kimi K2 Thinking Moonshot AI | Q4_K_M | won't run | 580.5 GB | – | – | – | – |
| Kimi K2.5 Moonshot AI | Q4_K_M | won't run | 580.5 GB | – | – | – | – |
| MiniMax M2 MiniMax | Q4_K_M | won't run | 132.1 GB | – | – | – | – |
| MiniMax M2.1 MiniMax | Q4_K_M | won't run | 132.1 GB | – | – | – | – |
| MiniMax M2.5 MiniMax | MXFP4 | won't run | 118.6 GB | – | – | – | – |
| MiniMax M3 MiniMax | MXFP4 | won't run | 242.0 GB | – | – | – | – |
| Mistral Medium 3.5 128B Mistral AI | Q4_K_M | won't run | 73.9 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?
- Qwen 3.8 27B (Q8_0) fits fully in its memory at 8k context and generates at about 12 tok/s, an estimate from its memory bandwidth. 61 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 92 GB at 8k 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 8k context. See the best hardware for local AI at every budget.