HologramHologram
ModelsIndex Nov 20, 2026

Qwen3.8-27B-GGUF

7.1M6.2k

unsloth · 27.3B parameters · 480 GB · Apache 2.0

Address
unsloth/qwen3.8-27b-gguf
Get it
Status
Verified
Base model
…
Lineage
…
Unchanged from base
…
Trending
#17
Downloads, 30 days
7.1M
Weights
480 GB
Revision

Following the widespread community adoption of the Qwen3.5 and Qwen3.6 series, we are pleased to introduce Qwen3.8, the most capable generation in the Qwen open-model family to date.

At a glance

Parameters
27.3B
Architecture
Qwen 3.5
Context
256K tokens
Format
GGUF
License
Apache 2.0Commercial use
Base model
Quantized from Qwen/Qwen3.8-27B
Released
Aug 2026
Updated
Nov 2026
Likes
5,773
Downloads, all time
24,815,500

Architecture

Layers
64
Hidden size
5,120
Attention
24 heads, grouped query, 4 KV heads
Vocabulary
248,320
Positions
262,144
Tied embeddings
No
Vision encoder
qwen3_5_vision

Family

Models built on Qwen3.8-27B-GGUF.

Run it

Pinned to the indexed revision.

llama-server -hf unsloth/Qwen3.8-27B-GGUF

Through the hub: the same tools, each file from a source that is up (Hugging Face, ModelScope, IPFS), at this revision. The last line checks every file against the hash the tree names.

export HF_ENDPOINT=https://gethologram.ai
hf download unsloth/Qwen3.8-27B-GGUF --quiet && hf cache verify unsloth/Qwen3.8-27B-GGUF

Spaces

Used in 25 Spaces.

Read the full model card

Read our How to Run Qwen3.8-27B Guide!

[Unsloth Dynamic 3.1](https://unsloth.ai/docs/basics/dynamic-3.0-ggufs) achieves superior accuracy & outperforms other leading quants.




[![](https://github.com/unslothai/unsloth/raw/main/images/unsloth%20new%20logo.png)](https://github.com/unslothai/unsloth/)
[![](https://github.com/unslothai/unsloth/raw/main/images/Discord%20button.png)](https://discord.gg/unsloth)
[![](https://raw.githubusercontent.com/unslothai/unsloth/refs/heads/main/images/documentation%20green%20button.png)](https://unsloth.ai/docs/models/qwen3.8)



[Nov 11 update:](https://unsloth.ai/docs/models/qwen3.8) Dynamic v3.1 GGUF is here with 5-10% more average accuracy than other frontier quants. Benchmarks: Terminal Bench, DeepSWE, LiveCodeBench & more! [Details](https://unsloth.ai/docs/models/qwen3.8)
Run and fine-tune Qwen3.8 in [Unsloth Desktop](https://unsloth.ai/docs/new/desktop) with Thinking toggles. [Download](https://unsloth.ai) for Mac, Windows and Linux. [GitHub repo](github.com/unslothai/unsloth)

qwen3.8 dynamic v3.1

See below for 4-bit Qwen3.8-27B run inside of Unsloth Desktop: qwen3.8 unsloth desktop


Qwen3.8-27B

Following the widespread community adoption of the Qwen3.5 and Qwen3.6 series, we are pleased to introduce Qwen3.8, the most capable generation in the Qwen open-model family to date.

Built on the architectural foundation of Qwen3.5, Qwen3.8 delivers substantial gains across coding, professional work, research, and long-horizon agentic tasks. Qwen3.8-27B brings these advances to a compact, deployment-friendly dense model: a native vision-language model that understands images and videos, with flexible thinking control, designed to carry complex, multi-step tasks through to completion with greater reliability.

Qwen3.8 Highlights

Qwen3.8-27B features the following enhancements:

  • Core Capabilities: Comprehensive improvements across coding, professional work, research, and long-horizon agentic tasks.
  • Agent Execution: Stronger autonomous planning and better handling of environment feedback, leading to more reliable end-to-end task completion.
  • Downstream Compatibility: Broader support for popular harnesses and development tools, making it easier to integrate into your existing stack.
  • Flexible Thinking Control: Thinking mode is on by default and can be disabled per request; reasoning depth can be tuned with reasoning_effort, and reasoning context from historical messages is retained via preserve_thinking.
  • Vision-Language Understanding: Native support for image and video understanding, from STEM diagrams and documents to hour-scale videos.

Model Overview

  • Type: Causal Language Model with Vision Encoder
  • Training Stage: Pre-training & Post-training
  • Language Model
    • Number of Parameters: 27B
    • Hidden Dimension: 5120
    • Token Embedding: 248,320 (Padded)
    • Number of Layers: 64
    • Hidden Layout: 16 × (3 × (Gated DeltaNet → FFN) → 1 × (Gated Attention → FFN))
    • Gated DeltaNet:
      • Number of Linear Attention Heads: 48 for V and 16 for QK
      • Head Dimension: 128
    • Gated Attention:
      • Number of Attention Heads: 24 for Q and 4 for KV
      • Head Dimension: 256
      • Rotary Position Embedding Dimension: 64
    • Feed Forward Network:
      • Intermediate Dimension: 17,408
    • LM Output: 248,320 (Padded)
    • MTP (Multi-Token Prediction): trained with multiple steps
  • Context Length: 262,144 natively and extensible up to 1,000,000 tokens.

Best Practices

To achieve optimal performance, we recommend the following settings:

  1. Sampling Parameters: We suggest using the following sets of sampling parameters:

    • Thinking Mode: temperature=1.0, top_p=0.95, top_k=20, min_p=0.0, presence_penalty=0.0, repetition_penalty=1.0
    • Instruct (or non-thinking) mode: temperature=0.7, top_p=0.80, top_k=20, min_p=0.0, presence_penalty=1.5, repetition_penalty=1.0

    For supported frameworks, you can adjust the presence_penalty parameter between 0 and 2 to reduce endless repetition. However, using a higher value may occasionally result in language mixing and a slight decrease in model performance.

  2. Adequate Output Length: To optimize performance on agentic tasks, we recommend allocating sufficient output length to allow the model to generate detailed and comprehensive responses. For frameworks that support separate token limits for internal reasoning and final outputs, we suggest the following configuration within the 1M context length:

    • Reasoning Content: Set the maximum output length to 262,144 tokens.
    • Final Response: Set the maximum output length to 131,072 tokens.

    These settings provide the necessary capacity for complex reasoning while ensuring ample space for high-quality final deliverables.

  3. Processing Ultra-Long Texts: Qwen3.8-27B natively supports context lengths of up to 262,144 tokens. For long-horizon tasks where the total length (including both input and output) exceeds this limit, we recommend using RoPE scaling techniques to handle long texts effectively, e.g., YaRN.

  4. Long Video Understanding: To optimize inference efficiency for plain text and images, the size parameter in the released video_preprocessor_config.json is conservatively configured. It is recommended to set the longest_edge parameter in the video_preprocessor_config file to 469,762,048 (corresponding to 224k video tokens) to enable higher frame-rate sampling for hour-scale videos and thereby achieve superior performance. For example,

    {"longest_edge": 469762048, "shortest_edge": 4096}
    

Citation

If you find our work helpful, feel free to give us a cite.

@misc{qwen38,
    title = {{Qwen3.8-Max}: A New Bar for Coding and Cowork},
    url = {https://qwen.ai/blog?id=qwen3.8},
    author = {{Qwen Team}},
    month = {August},
    year = {2026}
}

Derived on Nov 12, 2026 from Hugging Face at revision 4af51b35, README.md , config.json .