Radeon AI Pro R9700 for local LLMs 20 measured results from 5 builds, each linked to its source.
Memory 32 GB per card Memory bandwidth not listed Typical EU price ≈€1,848 new (PowerColor Radeon AI PRO R9700 32 GB GDDR6, billiger.de , 2026-10-24)
Can it run? Qwen3.6 27B on Radeon AI Pro R9700 Qwen3.6 35B-A3B on Radeon AI Pro R9700 Qwen3.8 27B on Radeon AI Pro R9700 gpt-oss 20B on Radeon AI Pro R9700
All builds with the Radeon AI Pro R9700 All measured results (MTP, DFlash, DSpark and similar draft-and-verify methods give much higher tok/s on predictable text and are not comparable 1:1 with plain decode)
Model Build Ran on Quant Engine tok/s Details Qwen3.6 35B-A3B Radeon AI Pro R9700 (one card): Qwen3.6 35B-A3B from 4K to 256K context (AMDai lab rig) Radeon AI Pro R9700 Q4_K_M llama.cpp Vulkan 117.69 Qwen3.6 35B-A3B Radeon AI Pro R9700 (one card): Qwen3.6 35B-A3B from 4K to 256K context (AMDai lab rig) Radeon AI Pro R9700 Q5_K_M (24.68 GB VRAM) llama.cpp Vulkan 109.73 Qwen3.6 35B-A3B Radeon AI Pro R9700 (one card): Qwen3.6 35B-A3B from 4K to 256K context (AMDai lab rig) Radeon AI Pro R9700 Unsloth UD-Q5_K_XL, short context, f16 KV llama.cpp Vulkan 102.39 Qwen3.6 35B-A3B Radeon AI Pro R9700 (one card): Qwen3.6 35B-A3B from 4K to 256K context (AMDai lab rig) Radeon AI Pro R9700 UD-Q5_K_XL, 32K context llama.cpp Vulkan 90.52 Qwen3.6 35B-A3B Radeon AI Pro R9700 (one card): Qwen3.6 35B-A3B from 4K to 256K context (AMDai lab rig) Radeon AI Pro R9700 UD-Q5_K_XL, 128K context llama.cpp Vulkan 66.43 Qwen3.6 35B-A3B Radeon AI Pro R9700 (one card): Qwen3.6 35B-A3B from 4K to 256K context (AMDai lab rig) Radeon AI Pro R9700 UD-Q5_K_XL, 256K context llama.cpp Vulkan 44.4 Qwen3.8 27B Radeon AI Pro R9700: Qwen3.8 27B on llama.cpp, Ollama and vLLM (Vadi Taslim) Radeon AI Pro R9700 Q4_K_M + MTP speculative decoding, reasoning on llama.cpp Vulkan 57 ⚡ spec-decode (MTP) Qwen3.8 27B Radeon AI Pro R9700: Qwen3.8 27B on llama.cpp, Ollama and vLLM (Vadi Taslim) Radeon AI Pro R9700 Q4_K_M Ollama 30 Qwen3.8 27B Radeon AI Pro R9700: Qwen3.8 27B on llama.cpp, Ollama and vLLM (Vadi Taslim) Radeon AI Pro R9700 AWQ INT4, single request vLLM (radiance) 22 Qwen3.8 Flash-Next Strix Halo + Radeon AI Pro R9700 over OCuLink: Qwen3.8 Flash-Next at ~102 tok/s (custom engine) mixed / see build custom int4 experts / 6-bit trunk strixpro (MTP depth 3 + prompt lookup) 102 ⚡ spec-decode (MTP) Qwen3.8 Flash-Next Strix Halo + Radeon AI Pro R9700 over OCuLink: Qwen3.8 Flash-Next at ~102 tok/s (custom engine) mixed / see build custom int4 experts / 6-bit trunk strixpro, ~100K context 82 Qwen3.8 Flash-Next Strix Halo + Radeon AI Pro R9700 over OCuLink: Qwen3.8 Flash-Next at ~102 tok/s (custom engine) mixed / see build custom int4 experts / 6-bit trunk strixpro, ~980K context 71 Qwen3.8 Flash-Next Strix Halo + Radeon AI Pro R9700 over OCuLink: Qwen3.8 Flash-Next at ~102 tok/s (custom engine) mixed / see build GGUF llama.cpp (Strix Halo iGPU only) 29-34 gpt-oss 20B Intel Core Ultra 265K + Radeon AI Pro R9700 Radeon AI Pro R9700 llama.cpp 163.26 DeepSeek R1 Distill 14B legacy Intel Core Ultra 265K + Radeon AI Pro R9700 Radeon AI Pro R9700 llama.cpp 53.72 Llama 3.2 3B legacy Intel Core Ultra 265K + Radeon AI Pro R9700 Radeon AI Pro R9700 llama.cpp 195.91 Qwen3.6 27B legacy Radeon AI Pro R9700: Qwen3.6 27B on Vulkan, HIP and Ollama, with and without MTP Radeon AI Pro R9700 Q4_K_M + MTP draft head (draft-n-max 3), 64K context llama.cpp Vulkan 66.0 ⚡ spec-decode (MTP) Qwen3.6 27B legacy Radeon AI Pro R9700: Qwen3.6 27B on Vulkan, HIP and Ollama, with and without MTP Radeon AI Pro R9700 Q4_K_M, no MTP llama.cpp Vulkan 31.9 Qwen3.6 27B legacy Radeon AI Pro R9700: Qwen3.6 27B on Vulkan, HIP and Ollama, with and without MTP Radeon AI Pro R9700 Q4_K_M llama.cpp HIP (ROCm) 27.1 Qwen3.6 27B legacy Radeon AI Pro R9700: Qwen3.6 27B on Vulkan, HIP and Ollama, with and without MTP Radeon AI Pro R9700 Q4_K_M Ollama (HIP) 26.7
All Radeon AI Pro R9700 builds: see also all GPUs · best LLMs by memory size · download the dataset .
FAQ How fast is the Radeon AI Pro R9700 for local LLMs? The fastest plain-decode result we have is 195.91 tok/s on Llama 3.2 3B (, llama.cpp) in Intel Core Ultra 265K + Radeon AI Pro R9700 . Speed depends heavily on the model, quant and engine, so compare like with like in the table. RigBase lists 20 measured results from 5 builds.
How much memory does the Radeon AI Pro R9700 have? 32 GB per card.
Which models have been measured on the Radeon AI Pro R9700? DeepSeek R1 Distill 14B, Llama 3.2 3B, Qwen3.6 27B, Qwen3.6 35B-A3B, Qwen3.8 27B, Qwen3.8 Flash-Next, gpt-oss 20B.
Where do these numbers come from? Every row links to a build page with its original public source (benchmark logs, READMEs, forum posts). RigBase does not re-run the benchmarks. Results marked ⚡ use speculative decoding and are not comparable 1:1 with plain decode.
Ad · this space is available Sponsor this spot: reach people building local AI rigs. One clearly-labeled slot per page, no tracking scripts. Sponsor this spot → Not enough VRAM? Rent before you buy. Run Radeon AI Pro R9700-class models on a rented 24–96 GB cloud GPU by the hour and see the real speed before spending €1,000+.
Try it on Runpod Referral link: if you sign up with Google and add $10, Runpod gives you and RigBase bonus credit (terms ). It never affects our data.