readygpu

AMD Radeon AI PRO R9700 vs AMD Radeon PRO W7900

For running models locally, priced in Germany. At 128k context the first runs 50 models fully in memory, the second 60.

In short

The AMD Radeon PRO W7900 has more memory for models: 48 GB against 32 GB. At 128k context the AMD Radeon PRO W7900 runs 60 of the models we track fully in memory and the AMD Radeon AI PRO R9700 50; only the AMD Radeon PRO W7900 holds Qwen 3.5 122B-A10B. On the models both run fully in memory, the AMD Radeon PRO W7900 is about 1.3× as fast as the AMD Radeon AI PRO R9700 (estimated, median across models).

Updated . Speeds are estimates; how they are worked out.

AMD Radeon AI PRO R9700

2.049 €

Amazon

Add to build
AMD Radeon PRO W7900

Not priced here

Add to build

Specifications

SpecAMD Radeon AI PRO R9700AMD Radeon PRO W7900
GPURadeon AI PRO R9700Radeon PRO W7900
Memory32 GB GDDR648 GB GDDR6
Memory bus256-bit384-bit
Memory bandwidth640 GB/s864 GB/s
FP16 compute191 TFLOPS122.6 TFLOPS
SoftwareROCmROCm
Board power300 W295 W
PCIe generationGen 5Gen 4
PCIe lanesx16x16

What each runs

Estimated decode speed with the most precise quant that fits each one est.

ModelAMD Radeon AI PRO R9700AMD Radeon PRO W7900
Qwen 3.8 27B12tok/sUD-Q6_K13tok/sQ8_0
Qwen 3.5 122B-A10Boffload42tok/sUD-Q2_K_XL
Qwen 3.6 27B11tok/sQ6_K13tok/sQ8_0
Gemma 4 31B12tok/sQ4_K_M11tok/sQ8_0
Mistral Small 4 119Boffload53tok/sUD-Q2_K_XL
Qwen 3.6 35B-A3B43tok/sUD-Q5_K_M49tok/sQ8_0
Qwen 3.5 27B11tok/sQ6_K13tok/sQ8_0
Gemma 4 26B-A4B30tok/sQ8_040tok/sQ8_0
Qwen 3.5 35B-A3B41tok/sQ6_K49tok/sQ8_0
GLM 4.7 Flash19tok/sQ6_K23tok/sQ8_0
Devstral Small 2 24Boffload10tok/sQ8_0
Nemotron 3.5 Lightning 30B-A3B57tok/sUD-Q5_K_M68tok/sUD-Q6_K_XL
Kimi Linear 48B-A3B72tok/sQ4_K_M80tok/sQ6_K
gpt-oss 20B37tok/sMXFP450tok/sMXFP4
Qwen3 Coder 30B-A3B14tok/sQ4_K_M18tok/sQ8_0
Gemma 4 12B23tok/sQ8_031tok/sQ8_0
Qwen 3.5 9B25tok/sQ8_034tok/sQ8_0
Gemma 3 27B12tok/sQ5_K_M12tok/sQ8_0
Mistral Small 3.2 24Boffload10tok/sQ8_0
Qwen3 14B12tok/sQ4_K_M13tok/sQ8_0
Ministral 3 14B11tok/sQ5_K_M13tok/sQ8_0
DeepSeek R1 0528 Qwen3 8B13tok/sQ8_017tok/sQ8_0
Qwen3 8B13tok/sQ8_017tok/sQ8_0
Phi-4 14Boffload11tok/sQ8_0
Qwen 3.5 4B40tok/sQ8_054tok/sQ8_0
Ministral 3 8B13tok/sQ8_017tok/sQ8_0
Granite 4.2 8B11tok/sQ8_015tok/sQ8_0
Gemma 4 E4B29tok/sQ8_040tok/sQ8_0
LFM2.5 8B-A1B28tok/sQ8_038tok/sQ8_0
Llama 3.1 8B15tok/sQ6_K20tok/sQ6_K
Qwen2.5 0.5B164tok/sQ8_0222tok/sQ8_0
Qwen3.5 0.8B144tok/sQ8_0195tok/sQ8_0
Qwen3.5 2B115tok/sQ5_K_M155tok/sQ5_K_M
DeepSeek R1 Distill Qwen 1.5B62tok/sQ8_084tok/sQ8_0
Mellum2.1 12B A2.5B Thinking46tok/sQ8_062tok/sQ8_0
Spark X2.5 4B44tok/smlx-4bit60tok/smlx-4bit
MiniCPM5 2B42tok/sQ8_057tok/sQ8_0
FrogNano 4B 260939tok/sQ8_053tok/sQ8_0
LFM2.5 2.6B32tok/sQ8_044tok/sQ8_0
LFM2.5 VL 3B32tok/sQ8_044tok/sQ8_0
Kimi VL A3B Thinking 250629tok/sQ8_040tok/sQ8_0
Kimi VL A3B29tok/smlx-8bit39tok/smlx-8bit
GLM 4 9B 041426tok/smlx-5bit35tok/smlx-5bit
granite vision 4.1 4b25tok/sQ8_033tok/sQ8_0
granite 4.2 3b24tok/sQ8_033tok/sQ8_0
GLM 4.6V Flash23tok/sQ8_031tok/sQ8_0
DeepSeek R1 Distill Qwen 7B23tok/sQ8_030tok/sQ8_0
GLM 4.1V 9B Thinking21tok/smlx-8bit28tok/smlx-8bit
Ministral 3 3B Instruct 251220tok/sQ8_027tok/sQ8_0
LFM2.5 24B A2B17tok/sQ5_K_M22tok/sQ8_0
Muse Glimmer 30B14tok/sQ6_K14tok/sQ8_0
DeepSeek R1 Distill Llama 8B14tok/sQ8_019tok/sQ8_0
Ministral 3 8B Reasoning 251213tok/sQ8_017tok/sQ8_0
Qwen3.8 27B OBLITERATED12tok/smlx-4bit16tok/smlx-4bit
Ministral 3 14B Reasoning 251211tok/sQ5_K_M13tok/sQ8_0
DeepSeek R1 Distill Qwen 14Boffload12tok/sQ8_0
Devstral Small 2507offload10tok/sQ8_0
Magistral Small 2506offload10tok/sQ8_0
Magistral Small 2507offload10tok/sQ8_0
Magistral Small 2509offload10tok/sQ8_0
Qwen3 Coder Nextoffloadoffload
Nemotron 3 Super 120B-A12Boffloadoffload
gpt-oss 120Bwon't runoffload
GLM 4.5 Airwon't runoffload
Qwen3 Next 80B-A3Boffloadoffload
Qwen3 32Boffloadoffload
DeepSeek R1 Distill Qwen 32Boffloadoffload
granite 4.2 30boffloadoffload
DeepSeek V4 Flashwon't runwon't run
GLM 5.3 Flashwon't runwon't run
Qwen 3.8 Flash Nextwon't runwon't run
MiniMax M2.7won't runwon't run
Llama 3.3 70Bwon't runwon't run
DeepSeek R1won't runwon't run
DeepSeek R1 0528won't runwon't run
DeepSeek R1 Distill Llama 70Bwon't runwon't run
DeepSeek R1 Zerowon't runwon't run
DeepSeek V3won't runwon't run
DeepSeek V3 0324won't runwon't run
DeepSeek V3.1won't runwon't run
DeepSeek V3.1 Terminuswon't runwon't run
DeepSeek V3.2won't runwon't run
DeepSeek V4 Flash 0731won't runwon't run
DeepSeek V4 Flash Vision Expwon't runwon't run
Devstral 2 123B Instruct 2512won't runwon't run
GLM 4.5won't runwon't run
GLM 4.5Vwon't runwon't run
GLM 4.6won't runwon't run
GLM 4.6Vwon't runwon't run
GLM 4.7won't runwon't run
GLM 5won't runwon't run
GLM 5.1won't runwon't run
GLM 5.2won't runwon't run
GLM 5.3won't runwon't run
Kimi Dev 72Bwon't runwon't run
Kimi K2won't runwon't run
Kimi K2 Instruct 0905won't runwon't run
Kimi K2 Thinkingwon't runwon't run
Kimi K2.5won't runwon't run
MiniMax M2won't runwon't run
MiniMax M2.1won't runwon't run
MiniMax M2.5won't runwon't run
MiniMax M3won't runwon't run
Mistral Medium 3.5 128Bwon't runwon't run

Questions

Which is better for local AI, the AMD Radeon AI PRO R9700 or the AMD Radeon PRO W7900?
Choose the AMD Radeon PRO W7900 to run bigger models. Choose the AMD Radeon PRO W7900 for speed.
Which can run bigger models, the AMD Radeon AI PRO R9700 or the AMD Radeon PRO W7900?
At 128k context the AMD Radeon PRO W7900 runs 60 of the models we track fully in memory and the AMD Radeon AI PRO R9700 50; only the AMD Radeon PRO W7900 holds Qwen 3.5 122B-A10B.
Which is faster for local AI, the AMD Radeon AI PRO R9700 or the AMD Radeon PRO W7900?
On the models both run fully in memory, the AMD Radeon PRO W7900 is about 1.3× as fast as the AMD Radeon AI PRO R9700 (estimated, median across models).

More comparisons

More hardware for local AI