Model benchmarks

gemma-3b-it-expanded local LLM performance

As of November 2026, gemma-3b-it-expanded runs at up to 9.3 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).

LM StudioQ4_K_S
ShareRedditX

Model size

3B

Peak speed

9.3 tok/s

Average speed

9.3 tok/s

Avg PP

386.6 tok/s

Min memory

2.7 GB

Max context

16,384 tokens

Avg output / run

2,224 tokens

Avg runtime / run

4m 20s

Avg quality

12.5

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)
Overall10.912.514.1
Agent Workflow10.713.616.6
Code Generation0.00.50.9
Role Play & Narrative18.821.223.6
Research & Analysis9.314.820.3

Performance by hardware and tool

Every hardware/tool/quantization combination gemma-3b-it-expanded has been benchmarked on, ranked by peak token generation speed. Last updated November 2026.

HardwareToolQuantBest tok/sAvg tok/sMemoryContextQualityRuns
NVIDIA GeForce GTX 1050LM StudioQ4_K_S9.3 tok/s9.3 tok/s2.7 GB16,384 tokens12.52

Benchmark runs

All 2 gemma-3b-it-expanded runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.

Frequently asked questions

Is gemma-3b-it-expanded good for coding?
In our benchmarks, gemma-3b-it-expanded scores 0.5/100 for coding. It runs at about 9.3 tok/s, so if you want more speed, byteshape/Qwen3.8-27B-GGUF:IQ4_XS-ASCII-Condensed is faster (~42.0 tok/s) and still scores well for coding (86.3/100).
Is gemma-3b-it-expanded good for agentic (tool-using) tasks?
In our benchmarks, gemma-3b-it-expanded scores 13.6/100 for agentic workflows. It runs at about 9.3 tok/s, so if you want more speed, muse-glimmer:latest is faster (~33.3 tok/s) and still scores well for agentic workflows (91.6/100).
How fast is gemma-3b-it-expanded for local inference?
Across 2 community benchmark runs, gemma-3b-it-expanded reaches up to 9.3 tok/s and averages 9.3 tok/s, with the fastest results on NVIDIA GeForce GTX 1050.
How much memory does gemma-3b-it-expanded need?
The leanest observed configuration used about 2.7 GB of memory (quantizations tested: Q4_K_S).
Which tools have been used to run gemma-3b-it-expanded?
Benchmarks were submitted using LM Studio. Results are community-contributed and updated as new runs arrive.