RTX 4070

NVIDIA · 12 GB · 504 GB/s

Top 3 recipes

Recommended

Qwen3.8-27B

EXL3 2.2 bpwTabbyAPI32K context

Best for Coding agents, long documents, everyday assistant work.

54 tok/sdecode

681 tok/sprefill

32Kcontext

✓ Thinks✓ Tools✓ Long context✓ Vision

Tested on this card · Sep 26, 2026

Fastest

Qwen3.5-9B

EXL3 4 bpwTabbyAPI128K context

Best for Fast answers and light agent work on 8–12 GB cards.

104 tok/sdecode

1,449 tok/sprefill

128Kcontext

✓ Thinks✓ Tools✓ Long context

Tested on this card · Sep 25, 2026

Another family

Gemma 4 12B

EXL3 4 bpwTabbyAPI64K context

Best for Vision, documents and quick chat.

55 tok/sdecode

1,211 tok/sprefill

64Kcontext

✓ Thinks✓ Tools✓ Long context✓ Vision

Tested on this card · Sep 26, 2026