You can deploy this model and start earning money today!

AtomicChat/
Qwen3.5-4B-GGUF

Name: Qwen3.5-4B-GGUF
Availability: InStock

Views

No views yet

Base model: Qwen/Qwen3.5-4B

Qwen3.5 4B, self-quantized to GGUF by Atomic Chat. Built straight from Qwen's original weights with a per-tensor importance matrix, so this is not a repack of somebody else's files. Runs fully offline.

Highlights

4.7B parameters: the weights this repo quantizes.
Context length: 262,144 tokens (256K), as published by Qwen.
32 layers: Dense decoder.
Modalities: the base model handles Text, Image; this repo ships text-only quants, it carries no vision projector.
Full imatrix ladder: every quant is calibrated with an importance matrix.
Unified Vision-Language Foundation: Early fusion training on multimodal tokens achieves cross-generational parity with Qwen3 and outperforms Qwen3-VL models across reasoning, coding, agents, and visual understanding benchmarks.
Efficient Hybrid Architecture: Gated Delta Networks combined with sparse Mixture-of-Experts deliver high-throughput inference with minimal latency and cost overhead.

[!NOTE] These GGUFs are self-quantized from the original weights, not a repack. The importance matrix keeps low-bit quants closer to the full-precision model.

[!IMPORTANT] Always pass --jinja so the Qwen3.5 4B chat template is applied. Without it the model can emit malformed turns.

Model Overview

Property	Value
Base model	`Qwen/Qwen3.5-4B`
Parameters	4.7B
Layers	32
Context length	262,144 tokens (256K)
Vocabulary	248,320
Modalities	Text, Image in the base model; text only in this repo, it ships no vision projector
Architecture	Dense decoder, 16 attention heads over 4 KV heads, `Qwen3_5ForConditionalGeneration`
This repo	GGUF quants (imatrix). Quants: `Q2_K`, `IQ3_M`, `Q3_K_M`, `Q3_K_L`, `IQ4_XS`, `Q4_K_S`, `Q4_K_M`, `UD-Q4_K_XL`, `Q5_K_S`, `Q5_K_M`, `Q6_K`, `Q8_0`

Scores are Qwen's published results for the base Qwen/Qwen3.5-4B, not our own measurements. Quantization preserves the large majority of this; Q4_K_M and up stay close to full precision.

Choosing a quant

Quant	Size	Notes
`Q2_K`	1.9 GB	Smallest K-quant. Minimal RAM, clear quality drop.
`IQ3_M`	2.2 GB	Beats Q3 at a similar size thanks to imatrix. Best low-RAM pick.
`Q3_K_M`	2.3 GB	Low quality but usable.
`Q3_K_L`	2.4 GB	A step above Q3_K_M.
`IQ4_XS`	2.5 GB	Excellent quality for size. Recommended low-bit.
`Q4_K_S`	2.6 GB	Compact 4-bit, fast.
`Q4_K_M`	2.7 GB	Recommended default. Best balance of size, speed and quality.
`UD-Q4_K_XL`	2.9 GB	Dynamic. Embeddings and output kept at Q8_0 for higher quality at a Q4 footprint.
`Q5_K_S`	3.0 GB	Higher quality, slightly more compact than Q5_K_M.
`Q5_K_M`	3.1 GB	Higher quality, low loss.
`Q6_K`	3.5 GB	Near lossless, noticeably lighter than Q8_0.
`Q8_0`	4.5 GB	Effectively lossless, reference quality.

[!TIP] Pick the largest file that fits your (V)RAM with room for context. Q4_K_M or UD-Q4_K_XL is the sweet spot for most setups; Q6_K or Q8_0 for maximum fidelity.

Get started

Run Qwen3.5 4B locally with:

Atomic Chat: the easiest path. Open the app, search AtomicChat/Qwen3.5-4B-GGUF, pick a quant, hit Use this model.
llama.cpp: llama-server -hf AtomicChat/Qwen3.5-4B-GGUF:Q4_K_M --jinja -c 8192
Ollama: ollama run hf.co/AtomicChat/Qwen3.5-4B-GGUF:Q4_K_M
LM Studio / Jan: search the repo id, download any quant.

Best practices

Parameter	Value
temperature	1.0
top_p	0.95
top_k	20
min_p	0.0
repetition_penalty	1.0

Qwen's recommended sampling configuration for Qwen/Qwen3.5-4B.

Run in llama.cpp

bash

git clone https://github.com/ggml-org/llama.cpp
cmake llama.cpp -B llama.cpp/build -DBUILD_SHARED_LIBS=OFF -DGGML_CUDA=ON
cmake --build llama.cpp/build --config Release -j --target llama-cli llama-server

bash

./llama.cpp/build/bin/llama-server \
    -hf AtomicChat/Qwen3.5-4B-GGUF:Q4_K_M \
    --jinja -ngl 99 -c 8192 -fa on

How these were made

Download Qwen/Qwen3.5-4B (original weights).
Convert to f16 GGUF with llama.cpp.
Build an importance matrix over our calibration corpus.
Quantize the ladder with --imatrix.
UD-Q4_K_XL additionally pins the token-embedding and output tensors to Q8_0.

License

Original model by Qwen, released under the Apache 2.0 license. Full terms: Apache 2.0. Quantized by Atomic Chat.