Gemma 3 1B

Name: Gemma 3 1B
Author: Google

Google · Gemma 3 · released 2025-03-12 · data checked 2026-11-01 Needs verification

Gemma 3 1B is a 1B-parameter dense model from Google (United States), released 2025-03-12 under the Gemma license. It scores 6.1 on the Small LM Index, #21 of 21 tiny models. It needs about 1.2 GB of memory at 4-bit with an 8K-token context.

Small LM Index

6.1

#21 of 21 scored tiny models

Likely 4.6–7.9 · from 3 of 6 core benchmarks

Parameters

Dense

Price per 1M tokens

—

No hosted API on OpenRouter; self-host

Context window

32K

tokens

Benchmarks

Each bar shows this model's score. The tick marks the best tiny model on that benchmark. Sources: 3 reported by Google. How we label sources.

MMLU-Pro Lab-reported Knowledge14.7

Class best: Qwen3.5 4B 79.1

GPQA Diamond Lab-reported Knowledge19.2

Class best: Nanbeige4.2 3B 87.4

LiveCodeBench Lab-reported Coding · LiveCodeBench v51.9

Class best: LFM3 1.7B 73.5

Not reported: estimated

What the index model expects from this model's other scores. Not a measurement.

Humanity's Last Exam≈ 0

AIME≈ 1

SWE-bench Verified≈ 1

Specifications

Lab: Google (US · United States)
Architecture: Dense transformer
Total parameters: 1B
Active parameters: All (dense)
Size class: Tiny
Context window: 32K tokens
Input modalities: text
Reasoning: Instruct (no thinking)
License: Gemma
Openness: Open weights
Memory to run: ~1.2 GB at 4-bit, 8K context (more sizes)
Hugging Face: 212.6K downloads (30 days) · 1.3K likes

Weights on Hugging Face ↗

Compare with

Gemma 3 1B vs EXAONE 4.0 1.2B (1.3B, index 36.1)
Gemma 3 1B vs LFM2.5 1.2B Thinking (1.2B, index 30.1)
Gemma 3 1B vs LFM2.5 1.2B Instruct (1.2B, index 20.6)
Gemma 3 1B vs LFM2.5 350M (350M, index 12.6)
Gemma 3 1B vs Qwen3.5 0.8B (900M, index 10.7)

Sources

https://huggingface.co/google/gemma-3-1b-it

Something wrong here? Report it.

Run it locally

Uses google/gemma-3-1b-it-qat-q4_0-gguf (Q4_0). Swap the tag for another size: Q4_0.

Ollama

ollama run hf.co/google/gemma-3-1b-it-qat-q4_0-gguf:Q4_0

llama.cpp

llama-server -hf google/gemma-3-1b-it-qat-q4_0-gguf:Q4_0

Then open http://localhost:8080 for a chat window.

MLX (Mac)

pip install -U mlx-lm
mlx_lm.chat --model mlx-community/gemma-3-1b-it-qat-4bit

vLLM (GPU server)

vllm serve google/gemma-3-1b-it

Full-precision weights, for a server GPU.

Open in LM Studio

Memory needed: weights, context cache and runtime buffers.

Quantization	4K context	32K context
16-bit F16	2.5 GB	2.7 GB
8-bit Q8_0	1.6 GB	1.7 GB
4-bit Q4_K_M	1.2 GB	1.3 GB

Check it against your GPU, Mac or phone →