What is the best local model for coding on the RX 6700 6700 Xt 6750 Xt 6800m 6850m Xt?

On the RX 6700 6700 Xt 6750 Xt 6800m 6850m Xt, Qwen3.8-27B-GSQ-RCO-IQ3_XXS-mtp ranks first for coding in our benchmarks (85.1/100 at ~28.6 tok/s). Tiel-Coder-35B-A3B-MTP-UD-Q4_K_XL is faster (~33.9 tok/s) and still scores well (81.2/100) — 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 6700 6700 Xt 6750 Xt 6800m 6850m Xt?

On the RX 6700 6700 Xt 6750 Xt 6800m 6850m Xt, Ornith-1.5-9B-MTP-Q6_K ranks first for agentic workflows in our benchmarks (90.6/100 at ~44.4 tok/s). Ornith-1.5-9B-MTP-Q4_0_ROCMFP4_COHERENT is faster (~58.3 tok/s) and still scores well (88.2/100) — a good pick if you'd rather trade a little quality for speed.

Hardware Benchmarks

Best Local LLM Performance for RX 6700 6700 Xt 6750 Xt 6800m 6850m Xt

Compare RX 6700 6700 Xt 6750 Xt 6800m 6850m Xt benchmark results and see which models actually earn the best completed quality scores.

ShareReddit X

Hardware Overview

GPU-specific specifications for local LLM planning.

AMD Radeon RX 6700/6700 XT/6750 XT / 6800M/6850M XT

VRAM: 12GB

Recommended Models

Models we recommend for RX 6700 6700 Xt 6750 Xt 6800m 6850m Xt.

Fastest

Qwen3.8-27B-ROCMFP2

27B

5167.3 tok/s

Highest token generation speed at 5167.3 tok/s.

Best quality

Qwen3.8-27B-GSQ-RCO-IQ3_XXS-mtp

27B

28.6 tok/sQuality 84.5

Highest completed quality score at 84.5.

Frequently asked questions

Which models actually run best on the RX 6700 6700 Xt 6750 Xt 6800m 6850m Xt, by task, from community benchmark data.

What is the best local model for coding on the RX 6700 6700 Xt 6750 Xt 6800m 6850m Xt?: On the RX 6700 6700 Xt 6750 Xt 6800m 6850m Xt, Qwen3.8-27B-GSQ-RCO-IQ3_XXS-mtp ranks first for coding in our benchmarks (85.1/100 at ~28.6 tok/s). Tiel-Coder-35B-A3B-MTP-UD-Q4_K_XL is faster (~33.9 tok/s) and still scores well (81.2/100) — 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 6700 6700 Xt 6750 Xt 6800m 6850m Xt?: On the RX 6700 6700 Xt 6750 Xt 6800m 6850m Xt, Ornith-1.5-9B-MTP-Q6_K ranks first for agentic workflows in our benchmarks (90.6/100 at ~44.4 tok/s). Ornith-1.5-9B-MTP-Q4_0_ROCMFP4_COHERENT is faster (~58.3 tok/s) and still scores well (88.2/100) — a good pick if you'd rather trade a little quality for speed.

Benchmark Results

Completed quality-ranked benchmark results for AMD Radeon RX 6700/6700 XT/6750 XT / 6800M/6850M XT.

Open run page →