AMD Radeon AI Pro R9700

What hackers pick instead

• NVIDIA GeForce RTX 3090A used RTX 3090 is the go-to card for running local LLMs, with 24GB of VRAM, strong price-performance, and better speed than Apple hardware, though used prices have climbed.116recommend
• NVIDIA GeForce RTX 5090A top consumer GPU for very fast local AI inference and 4K gaming, but 32GB VRAM limits larger models and street prices run well above MSRP.102recommend
• NVIDIA RTX PRO 6000Recommended for local AI models and fine-tunes with 96GB VRAM and fast compute. Main caveat: very high price, and little gain over an RTX 5090 for inference.77recommend
• Nvidia GPUsPeople recommend Nvidia GPUs for top performance, mature software support and a strong CUDA ecosystem; critics point to high prices and closed-source drivers.67recommend
• AMD Radeon GPURecommended for its open-source Linux drivers, reliable kernel support, and good performance for the price. Critics say its software stack, especially for ML and AI work, lags behind Nvidia.48recommend

Every quote

The 40 most detailed of 95 comments. Each links to the original.

1. Radeon R9700 with 32 GB VRAM is relatively affordable for the amount of RAM and with llama.cpp it runs fast enough for most things.

   Recommends · Feb 2026 · in “Qwen3.5 122B and 35B models offer Sonnet 4.5 performance on local computers”

2. even a single R9700 runs the main models I am interested in at speeds that are acceptably faster than what I have here

   Recommends · Sep 2026 · in “Apple caught off guard by AI demand for Mac Mini and Mac Studio”

3. I have this setup, with 2x 32Gb cards. It's perfect for my needs, and cheaper than anything comparable from NV.

   Recommends · Apr 2026 · in “Taking on CUDA with ROCm: 'One Step After Another'”

4. they have been a pleasure to use. Drivers work out-of-the-box, and they are completely quiet when unused.

   Recommends · Mar 2026 · in “Qwen3.5 122B and 35B models offer Sonnet 4.5 performance on local computers”

5. For that kind of money I can run 2x AMD R9700 (on a server I already had) and get 150-250 tokens/sec

   Recommends · Oct 2026 · in “Microsoft Launches Surface Laptop Ultra, Surface RTX Spark Dev Box”

6. 2x AMD R9700 that gets 260+ tokens/sec to start and slows down to ~110 tokens/sec over 128k context

   Recommends · Sep 2026 · in “M5 Ultra Mac Studio Review”

7. have actually gotten more capable through community efforts at optimization, since their launch

   Mixed · Oct 2026 · in “Anthropic S-1”

8. LLM's run great on it, it's happily running gemma4 31b at the moment and I'm quite impressed.

   Mixed · Apr 2026 · in “Taking on CUDA with ROCm: 'One Step After Another'”

9. I have a 32GB R9700, which is much cheaper than a 5090, and runs 27B dense models at ~20 t/s

   Mixed · Apr 2026 · in “Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model”

10. An R9700 is $1350 and can get 100 TPS running Qwen3.6-35B-A3B Q5 with 130k context window

    Recommends · Jun 2026 · in “RTX 5080 and RTX 3090 Setup: 80 Tok/s on Qwen 3.6 27B Q8”

11. at least in ComfyUI and local LLMs I'm finding support for AMD has gotten really good.

    Recommends · Sep 2026 · in “Jeff – Jev-compatible 0.8B decision models, trained at home, ~30 ms”

12. can run a dozen or so sessions of qwen 3.8 flash next at once at > 100tps each

    Recommends · Oct 2026 · in “Anthropic Cracking Down on Video Game Rewrites”

13. It's crazy how price efficient the r9700 is compared to the Nvidia cards.

    Recommends · Jun 2026 · in “Qwen 3.6 27B is the sweet spot for local development”

14. I can easily run Qwen3.6 35B-A3B with Q5_K_M with a 260k+ context window

    Recommends · Jun 2026 · in “Nvidia RTX Spark”

15. it is great llm hardware because it has 32Gb for a reasonable price

    Recommends · Mar 2026 · in “Can I Run AI locally?”

16. I run dual R9700 boxes those idle at 150W Because of a Ryzen AM5

    Mixed · Sep 2026 · in “Best LLM for every budget, updated daily”

17. an AMD r9700 pro has ~32GB of VRAM for much cheaper than a 5090

    Recommends · Jul 2026 · in “Kimi K2.7 Code is generally available in GitHub Copilot”

18. R9700s do really well at low wattage and underclocking as well

    Recommends · Jun 2026 · in “Ask HN: Has anyone replaced Claude/GPT with a local model for daily coding?”

19. definitely looks interesting for the price compared to NVidia

    Recommends · Jun 2026 · in “Ask HN: Has anyone replaced Claude/GPT with a local model for daily coding?”

20. over a year after the release of "pro AI" cards like the r9700, they had basic kernels lacking in vllm

    Criticizes · Aug 2026 · in “DiffusionGemma Technical Report”

21. Stock vLLM runs so slowly on these cards compared with vLLM forks like Radiance

    Criticizes · Sep 2026 · in “Speculative Decoding in vLLM on AMD GPUs”

22. I would advise against if you want to run things like vllm.

    Criticizes · Apr 2026 · in “Taking on CUDA with ROCm: 'One Step After Another'”

23. Probably better to buy the Radeon AI Pro R9700 for $1400, which will be faster, supported for many years, and has a fan already.

    Recommends · Jul 2026 · in “Benchmarking 15 "E-Waste" GPUs with Modern Workloads”

24. A couple of new GPUs, like Radeon AI Pro 9700 at $1400 each, probably gets you fast enough for comfortable interactive use

    Recommends · Aug 2026 · in “AirLLM 70B inference with single 4GB GPU”

25. Highly recommend if you want to play with something that doesn't involve NVidia and/or unobtanium-class hardware.

    Recommends · Jun 2026 · in “GLM 5.2 Is Out”

26. I just racked some consumer GPUs in my garage (6x r9700s), cranking 24/7 pumping 20M+ tokens a day

    Recommends · Sep 2026 · in “Pause OpenAI Now”

27. I build operating systems and fix compilers with them and reverse engineer anything I want.

    Recommends · Oct 2026 · in “Anthropic Cracking Down on Video Game Rewrites”

28. it could obviously run with higher context limits and bigger models due to the extra vram

    Recommends · Jun 2026 · in “Qwen 3.6 27B is the sweet spot for local development”

29. I paid $1350 and threw an R9700 in an existing machine. That's a 4 month pay off or so.

    Recommends · Sep 2026 · in “Show HN: Sunk Cost – How long until a local LLM rig pays for itself?”

30. I've had a R9700 and it'd run the lower quants (4-5 bit?) of Qwen 3.5 27B reasonably

    Recommends · Aug 2026 · in “Ask HN: What can I do with a single R9700?”

31. Insert 2-4 $1200 r9700 GPUs in a Linux 7.0.0+ machine with 64GB+ of DDR4-5 memory

    Recommends · Jun 2026 · in “Three of our worst VC stories”

32. for the same money you’re far better off getting an RTX 6000, 5090, or R9700.

    Recommends · Jul 2026 · in “Nobody knows what a used GPU cluster is worth”

33. basically the professional version of the RX 9070XT but with 32GB of memory

    Recommends · Aug 2026 · in “Apple caught off guard by AI demand for Mac Mini and Mac Studio”

34. I can feed it sensitive data all day and not be worried where it's going.

    Recommends · Sep 2026 · in “Show HN: Sunk Cost – How long until a local LLM rig pays for itself?”

35. I'd much rather have an RTX 5090, an Arc B70 Pro, or an AMD AI PRO R9700

    Recommends · Apr 2026 · in “Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model”

36. The most cost-effective local option right now, I think, is dual R9700.

    Recommends · Sep 2026 · in “Notes on gotchas while migrating 35kb preprompts from Opus to self-hosted Ollama”

37. Or you could get a 9700 32GB or an Arc B70 for well under $2k, today.

    Recommends · Sep 2026 · in “M5 Ultra Mac Studio Review”

38. the R9700 would be the best of the two (compared to the Arc Pro B70)

    Recommends · Aug 2026 · in “Ask HN: What can I do with a single R9700?”

39. There is a good and cheap alternative: R9700 32GB + Qwen 3.5 27B

    Recommends · Jun 2026 · in “Why current LLM costs are not sustainable”

40. I'd recommend current Radeon AI Pro 9700 cards over the V620 now

    Recommends · Aug 2026 · in “We Have AI at Home Chapter 1: A Box of Scraps”