Fastest
gemma-4-26b-a4b
26B
199.7 tok/s
Highest token generation speed at 199.7 tok/s.
Hardware Benchmarks
Best Local LLM Performance for 2× Ai Pro R9700
Compare 2× Ai Pro R9700 benchmark results and see which models actually earn the best completed quality scores.
Hardware Overview
GPU-specific specifications for local LLM planning.
- VRAM
- 64GB
- Architecture
- Multi-GPU
Recommended Models
Models we recommend for 2× Ai Pro R9700.
Best quality & Largest context tested
qwen3.8-27b
27B
75.3 tok/sQuality 90.1Context 65,536 tokens
Highest completed quality score at 90.1. Largest context window tested on this hardware, at 65,536 tokens. Its memory wasn't measured, so this doesn't show the full window fits.
Models on the 2× Ai Pro R9700
Every model the community ran on this hardware. Up is better answers, right is faster. Time to finish counts every token the model wrote, thinking included, so a model that talks a lot ends up further left.
Best trade-off: quality 60+, and nothing else here is both faster and betterOther models · dot size = number of runs
Medians over 20 fully rated runs of 10 models. Each quantization is its own model.
Frequently asked questions
Which models actually run best on the 2× Ai Pro R9700, by task, from community benchmark data.
- What is the best local model for coding on the 2× Ai Pro R9700?
- On the 2× Ai Pro R9700, swift-1.5-27b ranks first for coding in our benchmarks (84.8/100 at ~96.8 tok/s). qwen3.8-27b-turbo is faster (~126.5 tok/s) and still scores well (84.0/100) — a good pick if you'd rather trade a little quality for speed.
- What is the best local model for agentic workflows on the 2× Ai Pro R9700?
- On the 2× Ai Pro R9700, qwen3.8-27b-full ranks first for agentic workflows in our benchmarks (91.3/100 at ~129.4 tok/s). swift-1.5-27b-turbo is faster (~158.6 tok/s) and still scores well (88.4/100) — a good pick if you'd rather trade a little quality for speed.
Benchmark Results
Completed quality-ranked benchmark results for 2× AMD Radeon AI PRO R9700.