Unsloth Dynamic 2.0 Quants

New 2.0 version of our Dynamic GGUF + Quants. Dynamic 2.0 achieves superior accuracy & SOTA quantization performance.

Gemma 4

Gemma 4 is Google's new model family including including E2B, E4B, 26B-A4B, and 31B.

Unsloth Diffusion GGUFs

Find GGUFs and other variants of diffusion based models like Qwen-Image and FLUX.

gpt-oss

OpenAI's gpt-oss-20b and gpt-oss-120b is here! The powerful open models are available in GGUF, original & 4-bit formats.

Qwen3-VL

Qwen's new multimodal vision models in GGUF, safetensor, and dynamic Unsloth formats.

DeepSeek-V3.1

DeepSeek's new 3.1 update to their V3 models!

Embedding Models

Run or fine-tune embedding models with Unsloth.

Ministral 3

Mistral Ministral 3: new multimodal models in Base, Instruct, and Reasoning variants, available in 3B, 8B, and 14B sizes.

Gemma 3n

Google Gemma 3n models, all versions including Dynamic GGUF, 4-bit, 16-bit and formats!

Phi-4 (All Versions)

Microsoft's Phi-4 models including Reasoning + Reasoning Plus & mini. Includes Dynamic 2.0 GGUF, 4-bit & 16-bit versions. Includes Unsloth's bug fixes

Deepseek V3 (All Versions)

Deepseek-V3-0324 and V3 - available in original, and Dynamic GGUF formats, with support for 2-8-bit quantized versions.

Mistral Small 3 (All Versions)

A collection of Mistral's new Small 3.2 and 3 models including GGUF, 4-bit and more!

Llama 3.3 (All Versions)

Meta's new Llama 3.3 (70B) model in all formats. Includes GGUF, 4-bit bnb and original versions.

Qwen QwQ-32B Collection

Qwen's reasoning models including QwQ (32B) & QVQ (72B) in formats: GGUF, dynamic 4-bit and 16-bit original versions.

Llama 3.2 Vision

Meta's Llama 3.2 vision models 11B and 90B. Include 4-bit bnb and original versions.

Llama 3.1 Collection

Meta's Llama 3.1 models including 8B, 70B, 405B. Includes 4-bit bnb and original versions.

Load 4bit models 4x faster

Native bitsandbytes 4bit pre quantized models

Gemma 4 QAT

Gemma 4 QAT (Quantization-Aware Training) for 3x less memory use and near original accuracy.

Qwen3.6

Qwen3.5

Qwen3.5 is Qwen's new model family including Qwen3.5 Small: 0.8B, 2B, 4B, 9B and Qwen3.5 Medium: 35B-A3B, 27B, 122B-A10B and 397B-A17B.

Qwen3-Coder

The Qwen3-Coder models deliver SOTA advancements in agentic coding and code tasks. Includes Qwen3-Coder-Next.

Qwen3

Qwen's new Qwen3 models. In Unsloth Dynamic 2.0, GGUF, 4-bit and 16-bit Safetensor formats. Includes 128K Context Length variants.

DeepSeek R1 (All Versions)

DeepSeek-R1-0528 is here! The most powerful reasoning open LLM, available in GGUF, original & 4-bit formats. Includes Llama & Qwen distilled models.

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.

Granite 4.0

IBM's new Granite-4.0 models! Run Dynamic GGUFs or fine-tune with Unsloth.

Llama 4

Meta's new Llama 4 multimodal models, Scout & Maverick. Includes Dynamic GGUFs, 16-bit & Dynamic 4-bit uploads. Run & fine-tune them with Unsloth!

Unsloth 4-bit Dynamic Quants

Unsloths Dynamic 4bit Quants selectively skips quantizing certain parameters; greatly improving accuracy while only using <10% more VRAM than BnB 4bit

Text-to-Speech (TTS) models

A collection of 4-bit, Dynamic 4-bit and 16-bit voice models including Sesame-CSM, OpenAI's Whisper, Orpheus. Fine-tune them with Unsloth now!

Llama 3.2

Meta's new Llama 3.2 vision and text models including 1B, 3B, 11B and 90B. Includes GGUF, 4-bit bnb and original versions.

Qwen2.5-VL (All Versions)

All versions of Qwen2.5-VL including the new 32B version and 4-bit, 16-bit and more!

Vision/multimodal Models

Collection of the most popular vision models including Llama 3.2, LlaVa, Qwen2 VL, Pixtral, PaliGemma and more!

Qwen 2.5 Coder

Complete collection of Code-specific model series for Qwen2.5 in bnb 4bit, 16bit and GGUF formats.

Qwen 2.5

4bit Instruct Models