Fastest
Nex-N2.5-mini-official
Unknown
118.5 tok/s
Highest token generation speed at 118.5 tok/s.
Hardware Benchmarks
Best Local LLM Performance for RX 7800 Xt
Compare RX 7800 Xt benchmark results and see which models actually earn the best completed quality scores.
Hardware Overview
GPU-specific specifications for local LLM planning.
- VRAM
- 16GB
Recommended Models
Models we recommend for RX 7800 Xt.
Best quality
Qwen3.8-27B-GSQ-RCO-IQ3_S-mtp-ASCII-Condensed
27B
49.8 tok/sQuality 89.8
Highest completed quality score at 89.8.
Largest context
MiMo-V2.6-Distill-Qwen-9B-Q8_0
9B
51.8 tok/sContext 200,000 tokens
Largest context that fits in 16GB VRAM, at 200,000 tokens.
Frequently asked questions
Which models actually run best on the RX 7800 Xt, by task, from community benchmark data.
- What is the best local model for coding on the RX 7800 Xt?
- On the RX 7800 Xt, Qwen3.8-27B-Turbo-Brilliance-50X-RI-GSQ-RCO-IQ3_XXS-mtp-ASCII-Condensed ranks first for coding in our benchmarks (85.8/100 at ~35.9 tok/s). qwen3.8-flash-next-coder-iq1_m is faster (~42.2 tok/s) and still scores well (84.9/100) — a good pick if you'd rather trade a little quality for speed.
- What is the best local model for agentic workflows on the RX 7800 Xt?
- On the RX 7800 Xt, Qwen3.8-27B-TurboBrill-IQ3XXS-mtp ranks first for agentic workflows in our benchmarks (93.4/100 at ~32.3 tok/s). granite4.2:8b-q8_0 is faster (~51.3 tok/s) and still scores well (91.9/100) — a good pick if you'd rather trade a little quality for speed.
Benchmark Results
Completed quality-ranked benchmark results for AMD Radeon RX 7800 XT.