Model benchmarks

unsloth-ud-iq4_xs local LLM performance

As of November 2026, unsloth-ud-iq4_xs runs at up to 81.5 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).

Strata
ShareRedditX

Model size

Unknown

Peak speed

81.5 tok/s

Average speed

74.0 tok/s

Avg PP

490.4 tok/s

Min memory

n/a

Max context

512,000 tokens

Avg output / run

62,974 tokens

Avg runtime / run

14m 21s

Avg quality

85.1

Benchmark runs

2

GPUs tested

1

Quality by task

Average LLM-judged quality (0–100) with the run-to-run spread shown as a P5–P95 band, overall and for each benchmark task, across all 2 runs. The low and high columns show how much the judge’s score varies between runs, and need at least two runs to display.

TaskP5 (low)AvgP95 (high)
Overall81.885.188.4
Agent Workflow91.792.092.4
Code Generation57.267.577.9
Role Play & Narrative94.294.895.3
Research & Analysis84.186.188.2

Performance by hardware and tool

Every hardware/tool/quantization combination unsloth-ud-iq4_xs has been benchmarked on, ranked by peak token generation speed. Last updated November 2026.

HardwareToolQuantBest tok/sAvg tok/sMemoryContextQualityRuns
NVIDIA GeForce RTX 4090Strata—81.5 tok/s74.0 tok/sn/a512,000 tokens85.12

Benchmark runs

All 2 unsloth-ud-iq4_xs runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.

Frequently asked questions

Is unsloth-ud-iq4_xs good for coding?
In our benchmarks, unsloth-ud-iq4_xs scores 67.5/100 for coding. It runs at about 74.0 tok/s, so if you want more speed, turbo is faster (~124.1 tok/s) and still scores well for coding (85.3/100).
Is unsloth-ud-iq4_xs good for agentic (tool-using) tasks?
In our benchmarks, unsloth-ud-iq4_xs scores 92.0/100 for agentic workflows. It runs at about 74.0 tok/s, so if you want more speed, Qwen3.8-27B-IQ5R-V4-DFlash2-TP2-i8KV-r21 is faster (~133.3 tok/s) and still scores well for agentic workflows (91.8/100).
How fast is unsloth-ud-iq4_xs for local inference?
Across 2 community benchmark runs, unsloth-ud-iq4_xs reaches up to 81.5 tok/s and averages 74.0 tok/s, with the fastest results on NVIDIA GeForce RTX 4090.
Which tools have been used to run unsloth-ud-iq4_xs?
Benchmarks were submitted using Strata. Results are community-contributed and updated as new runs arrive.