What is the best local model for coding on the Titan RTX?

On the Titan RTX, ukisai/Swift-1.5-Qwen3.8-27B-GGUF:IQ4_XS ranks first for coding in our benchmarks (86.3/100 at ~44.0 tok/s).

What is the best local model for agentic workflows on the Titan RTX?

On the Titan RTX, Qwen3.8-Flash-Next ranks first for agentic workflows in our benchmarks (89.7/100 at ~17.5 tok/s). ukisai/Swift-1.5-Qwen3.8-27B-GGUF:IQ4_XS is faster (~44.0 tok/s) and still scores well (89.0/100) — a good pick if you'd rather trade a little quality for speed.

Hardware Benchmarks

Best Local LLM Performance for Titan RTX

Compare Titan RTX benchmark results and see which models actually earn the best completed quality scores.

ShareReddit X

Hardware Overview

GPU-specific specifications for local LLM planning.

NVIDIA TITAN RTX

VRAM: 24GB

Recommended Models

Models we recommend for Titan RTX.

Fastest & Best quality

ukisai/Swift-1.5-Qwen3.8-27B-GGUF:IQ4_XS

27B

44.0 tok/sQuality 86.6

Highest token generation speed at 44.0 tok/s. Highest completed quality score at 86.6.

Frequently asked questions

Which models actually run best on the Titan RTX, by task, from community benchmark data.

What is the best local model for coding on the Titan RTX?: On the Titan RTX, ukisai/Swift-1.5-Qwen3.8-27B-GGUF:IQ4_XS ranks first for coding in our benchmarks (86.3/100 at ~44.0 tok/s).
What is the best local model for agentic workflows on the Titan RTX?: On the Titan RTX, Qwen3.8-Flash-Next ranks first for agentic workflows in our benchmarks (89.7/100 at ~17.5 tok/s). ukisai/Swift-1.5-Qwen3.8-27B-GGUF:IQ4_XS is faster (~44.0 tok/s) and still scores well (89.0/100) — a good pick if you'd rather trade a little quality for speed.

Benchmark Results

Completed quality-ranked benchmark results for NVIDIA TITAN RTX.

Open run page →