What is the best local model for coding on the RX 6800 Xt?

On the RX 6800 Xt, google/gemma-4-12b ranks first for coding in our benchmarks (66.2/100 at ~32.4 tok/s).

What is the best local model for agentic workflows on the RX 6800 Xt?

On the RX 6800 Xt, google/gemma-4-12b ranks first for agentic workflows in our benchmarks (72.8/100 at ~32.4 tok/s).

Hardware Benchmarks

Name: AMD Radeon RX 6800 XT
Brand: AMD

Compare RX 6800 Xt benchmark results and see which models actually earn the best completed quality scores.

GPU-specific specifications for local LLM planning.

AMD Radeon RX 6800 XT

Models we recommend for RX 6800 Xt.

Fastest

14B

95.3 tok/s

Highest token generation speed at 95.3 tok/s.

Best quality & Largest context

12B

32.4 tok/sQuality 73.4Context 8.192 tokens

Highest completed quality score at 73.4. Largest context that fits in 15GB VRAM, at 8.192 tokens.

Which models actually run best on the RX 6800 Xt, by task, from community benchmark data.

What is the best local model for coding on the RX 6800 Xt?: On the RX 6800 Xt, google/gemma-4-12b ranks first for coding in our benchmarks (66.2/100 at ~32.4 tok/s).
What is the best local model for agentic workflows on the RX 6800 Xt?: On the RX 6800 Xt, google/gemma-4-12b ranks first for agentic workflows in our benchmarks (72.8/100 at ~32.4 tok/s).

Completed quality-ranked benchmark results for AMD Radeon RX 6800 XT.