Skip to content

Unsloth · Chat

Gemma 3 12B IT

UnslothChatGemini familyOpen weights. One of its sources says the weights are publicly available.
Market median price
$0.272 blended / 1M
Max context
131K
Listings
3
Providers
2

Evidence and sources

🤗 unsloth/gemma-3-12b-itReleased 12 Mar 2025
All rankings: #375 on NanoGPT recentlySources: Models.dev data, Hugging FacePrimary data from Models.dev from Hugging Face: release date

Advantages and trade-offs

Compared with 10 representative chat models

Advantages

Open weights available

The weights are public. 4 of the 10 peers that say either way have open weights.

We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 06:07 UTC.Prices come from the model's maker where they sell it, otherwise their official partner, otherwise the market median. This comparison uses 6 maker's prices; 1 market median price.

Input & output

as stated by the model's providers
Text
Text

Capabilities

Supported operations and provider-stated features

Prompt caching

Listings

 Aihubmixunsloth/gemma-3-12b-itSource: AIHubMix data docsNot published by the provider$0.20$0.80
 NanoGPTInference by KitaniRegions: GLBunsloth/gemma-3-12b-it · int4Source: NanoGPTNot published by the provider$0.2646cache read $0.1323$0.2646
 NanoGPTunsloth/gemma-3-12b-itSources: Models.dev data, NanoGPT131K$0.272cache read $0.136$0.272

These providers advertise this model. Researchers have found some resellers serving a different, cheaper model, so see how to check whether a provider meets your needs.

Benchmarks

No benchmark scores yet.

Around the web

  • Gemma 4 Fine-tuning Guide | Unsloth Documentation
    Introducing Unsloth Desktop: the first local app to run & train models ✨Read more LogoLogo …Introducing Unsloth Desktop: the first local app to run & train models ✨Read more LogoLogo `⌘Ctrl``k` RedditDiscordGitHubNewsletter For the complete documentation index, see llms.txt. This page is al…

    unsloth.ai

    Docs
  • Gemma 4 - How to Run Locally | Unsloth Documentation
    Gemma-4-26B-A4B runs on 18GB (4-bit) or 28GB (8-bit). Gemma-4-31B needs 20GB RAM (4-bit) or 34GB …Gemma-4-26B-A4B runs on 18GB (4-bit) or 28GB (8-bit). Gemma-4-31B needs 20GB RAM (4-bit) or 34GB (8-bit). You can now run all GGUFs, MLX and fine-tune Gemma 4 in Unsloth Studio (see right). QAT varia…

    unsloth.ai

    Docs
  • Gemma 4 QAT | Unsloth Documentation
    Gemma 4 QAT includes: E2B, E4B, 12B, 26B-A4B, and 31B. They are multimodal, hybrid-thinking models …Gemma 4 QAT includes: E2B, E4B, 12B, 26B-A4B, and 31B. They are multimodal, hybrid-thinking models that support 140+ languages and up to 256K context. Run Gemma 4 QATQAT Analysis Gemma-4-E2B QAT runs…

    unsloth.ai

    Docs
  • Gemma 3 - a unsloth Collection
    Hugging Face's logo ## Gemma 3 All versions of Google's new multimodal models including QAT …Hugging Face's logo ## Gemma 3 All versions of Google's new multimodal models including QAT in 1B, 4B, 12B, and 27B sizes. In GGUF, dynamic 4-bit and 16-bit formats. #### unsloth/gemma-3-270m-it-GGUF…

    huggingface.co

    Model card
  • Google - Gemma 4 now in Unsloth! · unslothai/unsloth · Discussion #4800 · GitHub
    Google releases Gemma 4 with four new models: E2B, E4B, 26B-A4B, 31B. You can now …Google releases Gemma 4 with four new models: E2B, E4B, 26B-A4B, 31B. You can now run and train the Gemma 4 models in Unsloth. Guide / Blog: The multimodal reasoning models are licensed under Apach…

    github.com

    Repository
  • Gemma 4 - a unsloth Collection
    Hugging Face's logo ## Gemma 4 Gemma 4 is Google's new model family including including …Hugging Face's logo ## Gemma 4 Gemma 4 is Google's new model family including including E2B, E4B, 26B-A4B, and 31B. #### unsloth/gemma-4-12b-it-GGUF Note GGUFs to run: #### unsloth/gemma-4-26B-A4B-it…

    huggingface.co

    Model card
  • unsloth/gemma-3-4b-it-GGUF · Hugging Face
    Gemma is a family of lightweight, state-of-the-art open models from Google, built from the same …Gemma is a family of lightweight, state-of-the-art open models from Google, built from the same research and technology used to create the Gemini models. Gemma 3 models are multimodal, handling text …

    huggingface.co

    Model card
  • Fine-tune Gemma 3 with Unsloth
    This means Unsloth is the only framework which works in float16 machines for Gemma 3! …This means Unsloth is the only framework which works in float16 machines for Gemma 3! This also means Colab Notebooks with free Tesla T4 GPUs also work! Gemma 3 (27B) finetuning fits with Unsloth in …

    unsloth.ai

  • Finetune Gemma 2 with Unsloth
    You can now finetune Gemma 2 (27b) with QLoRA to 9.7K context lengths (8K max …You can now finetune Gemma 2 (27b) with QLoRA to 9.7K context lengths (8K max supported) with Unsloth in a 40GB GPU, whilst HF+FA2 only allows 3K lengths. Unsloth can do 11K context lengths on a 24GB…

    unsloth.ai

  • Finetune Gemma with Unsloth
    You can now finetune Gemma 7b 2.43x faster than HF + Flash Attention 2 with …You can now finetune Gemma 7b 2.43x faster than HF + Flash Attention 2 with 57.5% less VRAM use. When compared to vanilla HF, Unsloth is 2.53x faster and uses 70% less VRAM. On 1x A100 80GB GPU, Unsl…

    unsloth.ai

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works