AMD · RDNA 4
Best AI Models for AMD Radeon AI PRO R9700 (32 GB)
Runs 27–31B and 35B MoE models well; 70B doesn't fit in 32 GB.
32 GB GDDR6640 GB/s191 TFLOPS FP16300 WRDNA 4 2025
| # | Model | Speed | llmrun Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | Qwen3.8 27BAlibabaHow to run → Q4_K_M · 17.9 GB · 22.0 t/s | 22.0 tok/s | Q4_K_M · 17.9 GB | |
| 2 | Qwen3.6 27BAlibaba Q4_K_M · 17.9 GB · 22.0 t/s | 22.0 tok/s | Q4_K_M · 17.9 GB | |
| 3 | Gemma 4 31B ITGoogle Q4_K_M · 21.7 GB · 18.8 t/s | 18.8 tok/s | Q4_K_M · 21.7 GB | |
| 4 | Gemma 4 26B A4B ITGoogle Q4_K_M · 16.5 GB · 76.0 t/s | 76.0 tok/s | Q4_K_M · 16.5 GB | |
| 5 | Q4_K_M · 19.3 GB · 70.6 t/s | 70.6 tok/s | Q4_K_M · 19.3 GB |
| # | Model | Speed | Coding Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | Qwen3.8 27BAlibabaHow to run → Q4_K_M · 17.9 GB · 22.0 t/s | 22.0 tok/s | Q4_K_M · 17.9 GB | |
| 2 | Gemma 4 31B ITGoogle Q4_K_M · 21.7 GB · 18.8 t/s | 18.8 tok/s | Q4_K_M · 21.7 GB | |
| 3 | Qwen3.6 35B A3BAlibaba Q4_K_M · 22.0 GB · 64.3 t/s | 64.3 tok/s | Q4_K_M · 22.0 GB | |
| 4 | Gemma 4 26B A4B ITGoogle Q4_K_M · 16.5 GB · 76.0 t/s | 76.0 tok/s | Q4_K_M · 16.5 GB | |
| 5 | Q4_K_M · 19.3 GB · 70.6 t/s | 70.6 tok/s | Q4_K_M · 19.3 GB |
| # | Model | Speed | Reasoning Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | Qwen3.8 27BAlibabaHow to run → Q4_K_M · 17.9 GB · 22.0 t/s | 22.0 tok/s | Q4_K_M · 17.9 GB | |
| 2 | Gemma 4 31B ITGoogle Q4_K_M · 21.7 GB · 18.8 t/s | 18.8 tok/s | Q4_K_M · 21.7 GB | |
| 3 | Qwen3.5 27BAlibaba Q4_K_M · 17.9 GB · 22.0 t/s | 22.0 tok/s | Q4_K_M · 17.9 GB | |
| 4 | Gemma 4 26B A4B ITGoogle Q4_K_M · 16.5 GB · 76.0 t/s | 76.0 tok/s | Q4_K_M · 16.5 GB | |
| 5 | Q4_K_M · 19.3 GB · 70.6 t/s | 70.6 tok/s | Q4_K_M · 19.3 GB |
| # | Model | Speed | Math Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | Qwen3.8 27BAlibabaHow to run → Q4_K_M · 17.9 GB · 22.0 t/s | 22.0 tok/s | Q4_K_M · 17.9 GB | |
| 2 | Qwen3.6 27BAlibaba Q4_K_M · 17.9 GB · 22.0 t/s | 22.0 tok/s | Q4_K_M · 17.9 GB | |
| 3 | Qwen3.6 35B A3BAlibaba Q4_K_M · 22.0 GB · 64.3 t/s | 64.3 tok/s | Q4_K_M · 22.0 GB | |
| 4 | Gemma 4 31B ITGoogle Q4_K_M · 21.7 GB · 18.8 t/s | 18.8 tok/s | overall | Q4_K_M · 21.7 GB |
| 5 | Q4_K_M · 19.3 GB · 70.6 t/s | 70.6 tok/s | Q4_K_M · 19.3 GB |
| # | Model | Speed ↓ | llmrun Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | Qwen3.5 4BAlibabaHow to run → Q4_K_M · 3.4 GB · 118 t/s | 118 tok/s | Q4_K_M · 3.4 GB | |
| 2 | GPT OSS 20BOpenAI Q4_K_M · 13.1 GB · 86.6 t/s | 86.6 tok/s | Q4_K_M · 13.1 GB | |
| 3 | Gemma 4 26B A4B ITGoogle Q4_K_M · 16.5 GB · 76.0 t/s | 76.0 tok/s | Q4_K_M · 16.5 GB | |
| 4 | Q4_K_M · 19.3 GB · 70.6 t/s | 70.6 tok/s | Q4_K_M · 19.3 GB | |
| 5 | Qwen3 30B A3B Thinking 2507Alibaba Q4_K_M · 19.4 GB · 66.9 t/s | 66.9 tok/s | Q4_K_M · 19.4 GB |
| # | Model | Speed | Vision Score | Quant · VRAM |
|---|---|---|---|---|
| 1 | Qwen3.8 27BAlibabaHow to run → Q4_K_M · 17.9 GB · 22.0 t/s | 22.0 tok/s | Q4_K_M · 17.9 GB | |
| 2 | Qwen3.6 27BAlibaba Q4_K_M · 17.9 GB · 22.0 t/s | 22.0 tok/s | Q4_K_M · 17.9 GB | |
| 3 | Gemma 4 31B ITGoogle Q4_K_M · 21.7 GB · 18.8 t/s | 18.8 tok/s | Q4_K_M · 21.7 GB | |
| 4 | Gemma 4 26B A4B ITGoogle Q4_K_M · 16.5 GB · 76.0 t/s | 76.0 tok/s | Q4_K_M · 16.5 GB | |
| 5 | MiniCPM V 4.7 35B A3BOpenBMB Q4_K_M · 22.1 GB · 69.6 t/s | 69.6 tok/s | Q4_K_M · 22.1 GB |
See all 346 models on the AI PRO R9700 →Best quant that fits · speeds estimated on this card
The AI PRO R9700 runs 27–31B and 35B MoE models well. RDNA 4 generation, released 2025.
GenerationRDNA 4released 2025
AI PRO R9700 strengths and weaknesses
Strengths
- Runs 35B MoE models: Qwen3.6 35B A3B at 64.3 tok/s in 22 GB
- Room for long context: Qwen3.8 27B at Q4 uses 17.9 GB, leaving 14.1 GB free
Weaknesses
- 70B models don't fit: Llama 3.3 70B Instruct needs 51.3 GB at Q4 vs 32 GB here
- Slow prompt reading: A 4K-token prompt on Qwen3.8 27B waits 5.4 s for the first token
- ROCm, not CUDA: No FP8 weights; MXFP4 only through llama.cpp, not vLLM
Output speed depends on memory bandwidth (640 GB/s). Grey rows need more than 32 GB.
- 9BQwen3.5 9B · Q4_K_M · 6.6 GB60.4tok/s
- 14BQwen3 14B · Q4_K_M · 11.9 GB40.4tok/s
- 27–31BQwen3.8 27B · Q4_K_M · 17.9 GB22.0tok/s
- 35B MoEQwen3.6 35B A3B · Q4_K_M · 22.0 GB64.3tok/s
- 70BLlama 3.3 70B Instruct · not at 4-bitneeds 51.3 GB
- 120B MoEGPT OSS 120B · not at 4-bitneeds 70.8 GB
Estimated speeds: Q4_K_M, 4K context, one user. How we estimate
All 1176 models on the AI PRO R9700
Featured onlyBest quant that fits, ordered by llmrun Score. Fit grades how well it runs here, S best. How ranking works
| Model | Quant | VRAM | Speed | Context | Score | Fit |
|---|---|---|---|---|---|---|
| Q4_K_M | 21.5 GB | 39.3 t/s | 16K | A75, Runs well | ||
| Q4_K_M | 17.3 GB | 25.9 t/s | 4K | A66, Runs well | ||
Vicuna 33B V1.3Chat | Q4_K_M | 21.8 GB | 17.6 t/s | 2K | B57, Decent | |
LFM2.5 1.2B BaseChat | Q4_K_M | 1.2 GB | 358.9 t/s | 128K | S100, Runs great | |
Llama Guard 3 1BChat | Q4_K_M | 1.7 GB | 302.4 t/s | 131K | S100, Runs great | |
Opt 6.7BChat | FP16 | 14.7 GB | 26.1 t/s | 2K | A67, Runs well | |
| Q4_K_M | 24.6 GB | 67.8 t/s | — | S85, Runs great | ||
| Q4_K_M | 4.8 GB | 145.5 t/s | — | S97, Runs great | ||
Qwen2.5 Coder 3B InstructChatCodeFunctions | Q4_K_M | 2.8 GB | 172.2 t/s | 33K | S98, Runs great | |
OpenCodeReasoning Nemotron 1.1 32BChatCodeReasoning | Q4_K_M | 24.3 GB | 18.7 t/s | 66K | B58, Decent | |
| BF16 | 18.6 GB | 23.1 t/s | 33K | B64, Decent | ||
Gemma 3 270M ITChat | Q4_K_M | 1.2 GB | 2133.3 t/s | — | S100, Runs great | |
| Q4_K_M | 0.8 GB | 662.1 t/s | 128K | S100, Runs great | ||
| Q4_K_M | 7.3 GB | 71.2 t/s | 131K | S86, Runs great | ||
II Medical 8BChat | Q4_K_M | 7.6 GB | 69.6 t/s | 41K | S86, Runs great |
Specs
- Architecture
- RDNA 4
- VRAM
- 32 GB GDDR6
- Memory bandwidth
- 640.0 GB/s
- FP16 (dense)
- 191.40 TFLOPS
- Stream processors
- 4,096
- TDP
- 300 W
- Released
- 2025-07-23
- MSRP
- $1,299
AI PRO R9700 FAQ
What is the largest model the AI PRO R9700 can run?
Qwen3.6 35B A3B (36.0B) is the largest at 4-bit or better: at Q4_K_M it uses 22.0 GB of 32 GB, and runs at about 64.3 tok/s.
Can the AI PRO R9700 run Qwen3.8 27B?
Yes. Qwen3.8 27B fits at Q4_K_M, using 17.9 GB of 32 GB, and it is the best model that runs well on this card.
Can the AI PRO R9700 run 70B models?
No. Llama 3.3 70B Instruct needs 51.3 GB at Q4, more than the 32 GB on this card.
How many tokens per second does the AI PRO R9700 give?
About 22.0 tok/s on Qwen3.8 27B at Q4_K_M, with the first token after 5.4 s on a 4K-token prompt.
What power supply does the AI PRO R9700 need?
Plan a 600 W PSU (2× the 300 W TDP).
Is 32 GB enough for coding and agents?
Yes for coding: Qwen3.8 27B, the best of the 10 coding models that fit, runs at about 22 tok/s. For agents, not yet: no agent-scored model fits in less than 96 GB.