# Qwen3-0.6B by Qwen: Open-Weight Model
Source: https://savrn.com/models/qwen3-0-6b
Markdown alternate of the page above; the site index is https://savrn.com/llms.txt

---

## Runs On

What it takes to serve Qwen3-0.6B (752M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.

| Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits |
| --- | --- | --- | --- | --- | --- |
| 16-bit | 1.5 GB | 1.8 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 |
| 8-bit | 0.8 GB | 0.9 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 |
| 4-bit | 0.4 GB | 0.5 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 |

Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the [SAVRN Index](https://savrn.com/ai-index/pricing/gpus), read Oct 23, 2026.

[Qwen3-0.6B on every accelerator the SAVRN Index prices, at every precision](https://savrn.com/models/qwen3-0-6b/gpus)

## SAVRN's Notes on Qwen3-0.6B

752M parameters paired with a 40,960-token context is the story here: a sub-billion-parameter model that reads whole documents. Memory runs 1.8 GB at 16-bit and 0.5 GB at 4-bit, so the cheapest setup we list — one 192 GB MI300X at $1.85 an hour — would hold more than a hundred copies; it rides beside larger workloads. At 29.6M downloads a month it leads every text-generation model in this part of the directory.

Apache 2.0 clears commercial use, modification and redistribution, notices kept and significant changes stated. The weights are one 1.5 GB safetensors file in bfloat16 over 28 layers. It derives from Qwen3-0.6B-Base, and the Qwen3 technical report (arXiv:2505.09388) sets out the design — start from the base if you plan your own post-training. No host on our Index quotes a per-token price, so the hourly card is the only cost signal.

SAVRN Research

## Model Card

By Qwen, published under apache-2.0, revision c1899de289a0.

### Qwen3 Highlights

Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support, with the following key features:

[Read the full model card (1,559 words)](https://savrn.com/models/qwen3-0-6b/card)

## Configuration

Architecture

Qwen3ForCausalLM

Context length (tokens)

40,960

Layers

28

Hidden size

1,024

Feed-forward size

3,072

Attention heads

16

Key/value heads

8

Head dimension

128

Vocabulary size

151,936

RoPE base

1,000,000

Stored precision

bfloat16

Model type

qwen3

## Identity and Version

Repository

Qwen/Qwen3-0.6B

Publisher

Qwen

Task

Text generation

Modality

Text

Library

transformers

Parameters

752M parameters

Languages

Not stated by the source

Revision

c1899de289a04d12100db370d81485cdf75e47ca

First published

2025-04-27

Last updated

2025-07-26

## Files and Weights

10 files, 1.5 GB in total. The weights are 1 file totalling 1.5 GB in safetensors.

Weights1 file · 1.5 GB

Configuration2 files · 965 B

Tokenizer4 files · 15.9 MB

Documentation2 files · 25.3 KB

Repository1 file · 1.6 KB

Every file

| File | Type | Size | SHA-256 |
| --- | --- | --- | --- |
| model.safetensors | Weights | 1.5 GB | f47f71177f32 |
| config.json | Configuration | 726 B | — |
| generation_config.json | Configuration | 239 B | — |
| LICENSE | Documentation | 11.3 KB | — |
| README.md | Documentation | 14.0 KB | — |
| .gitattributes | Repository | 1.6 KB | — |
| merges.txt | Tokenizer | 1.7 MB | — |
| tokenizer.json | Tokenizer | 11.4 MB | aeb13307a71a |
| tokenizer_config.json | Tokenizer | 9.7 KB | — |
| vocab.json | Tokenizer | 2.8 MB | — |

## License and Download

License

apache-2.0

Access

Open weights, no gate

Download size

1.5 GB

[Download from Qwen](https://modelscope.cn/models/Qwen/Qwen3-0.6B)

Released by Qwen through ModelScope. [Read the license](https://www.apache.org/licenses/LICENSE-2.0).

## Built From

- Derived from [Qwen/Qwen3-0.6B-Base](https://savrn.com/models/qwen3-0-6b-base)
- Described by [arXiv:2505.09388](https://savrn.com/papers/qwen3-technical-report)

## Memory Requirements

| Precision | Weights in memory |
| --- | --- |
| As published | 1.5 GB |
| 16-bit | 1.5 GB |
| 8-bit | 0.8 GB |
| 4-bit | 0.4 GB |

Weights only, from the published parameter count; the key-value cache and runtime add to this.

## Built on This Model

- Derived from[OmniVoice](https://savrn.com/models/omnivoice)
- Derived from[jina-reranker-v3](https://savrn.com/models/jina-reranker-v3)
- Derived from[Symbiotic-1B](https://savrn.com/models/symbiotic-1b)
- Derived from[Qwen3-0.6B-Distilled-30B-A3B-Thinking-SFT](https://savrn.com/models/qwen3-0-6b-distilled-30b-a3b-thinking-sft)
- Derived from[Qemma-sft](https://savrn.com/models/qemma-sft)
- Derived from[Qwen3-0.6B-Distilled-30B-A3B](https://savrn.com/models/qwen3-0-6b-distilled-30b-a3b)
- Quantized from[Qwen3-0.6B-int4](https://savrn.com/models/qwen3-0-6b-int4)
- Derived from[Qwen3-0.6B-int4](https://savrn.com/models/qwen3-0-6b-int4)
- Adapter of[Qwen3-0.6B-TLDR-Lora](https://savrn.com/models/qwen3-0-6b-tldr-lora)
- Derived from[Qwen3-0.6B-TLDR-Lora](https://savrn.com/models/qwen3-0-6b-tldr-lora)
- Adapter of[mini-Jev](https://savrn.com/models/mini-jev)
- Derived from[mini-Jev](https://savrn.com/models/mini-jev)
- Derived from[tiny-audio](https://savrn.com/models/tiny-audio)
- Quantized from[pulse-local-qwen3-0.6b](https://savrn.com/models/pulse-local-qwen3-0-6b)
- Derived from[pulse-local-qwen3-0.6b](https://savrn.com/models/pulse-local-qwen3-0-6b)
- Adapter of[qwen3-0.6b-cefc-rft-vietnamese-legal](https://savrn.com/models/qwen3-0-6b-cefc-rft-vietnamese-legal)
- Derived from[qwen3-0.6b-cefc-rft-vietnamese-legal](https://savrn.com/models/qwen3-0-6b-cefc-rft-vietnamese-legal)
- Derived from[social-jev-0.6B-preview](https://savrn.com/models/social-jev-0-6b-preview)
- Adapter of[qwen3-medical-qlora](https://savrn.com/models/qwen3-medical-qlora)
- Derived from[qwen3-medical-qlora](https://savrn.com/models/qwen3-medical-qlora)
- Quantized from[RK182X-LLM-Qwen3-0.6B](https://savrn.com/models/rk182x-llm-qwen3-0-6b)
- Derived from[RK182X-LLM-Qwen3-0.6B](https://savrn.com/models/rk182x-llm-qwen3-0-6b)
- Derived from[gpt-oss-20b-distill-Qwen3-0.6B](https://savrn.com/models/gpt-oss-20b-distill-qwen3-0-6b)
- Derived from[LightJev-0.6B-v0.1](https://savrn.com/models/lightjev-0-6b-v0-1)
- Derived from[OpenJev-0.6B](https://savrn.com/models/openjev-0-6b)
- Quantized from[Zircon-0.6B-v2-mlx](https://savrn.com/models/zircon-0-6b-v2-mlx)
- Derived from[Zircon-0.6B-v2-mlx](https://savrn.com/models/zircon-0-6b-v2-mlx)
- Adapter of[qwen3-06b-typed-decisions-cloud-pilot](https://savrn.com/models/qwen3-06b-typed-decisions-cloud-pilot)
- Derived from[qwen3-06b-typed-decisions-cloud-pilot](https://savrn.com/models/qwen3-06b-typed-decisions-cloud-pilot)
- Adapter of[qwen3-0.6b-cve-2025-3248-http-response-lora](https://savrn.com/models/qwen3-0-6b-cve-2025-3248-http-response-lora)
- Derived from[qwen3-0.6b-cve-2025-3248-http-response-lora](https://savrn.com/models/qwen3-0-6b-cve-2025-3248-http-response-lora)
- Quantized from[anascan-title-qwen3-0.6b-GGUF](https://savrn.com/models/anascan-title-qwen3-0-6b-gguf)
- Derived from[anascan-title-qwen3-0.6b-GGUF](https://savrn.com/models/anascan-title-qwen3-0-6b-gguf)
- Quantized from[Qwen3-0.6B-JNS](https://savrn.com/models/qwen3-0-6b-jns)
- Derived from[Qwen3-0.6B-JNS](https://savrn.com/models/qwen3-0-6b-jns)
- Derived from[noah-decide-v06-research](https://savrn.com/models/noah-decide-v06-research)
- Adapter of[qwen3-0.6b-math-tutor-4bit](https://savrn.com/models/qwen3-0-6b-math-tutor-4bit)
- Derived from[qwen3-0.6b-math-tutor-4bit](https://savrn.com/models/qwen3-0-6b-math-tutor-4bit)
- Adapter of[qwen3-0.6b-math-tutor-8bit](https://savrn.com/models/qwen3-0-6b-math-tutor-8bit)
- Derived from[qwen3-0.6b-math-tutor-8bit](https://savrn.com/models/qwen3-0-6b-math-tutor-8bit)

## Compare Qwen3-0.6B

- [gemma-3-1b-it vs Qwen3-0.6B](https://savrn.com/models/compare/gemma-3-1b-it-vs-qwen3-0-6b)
- [gpt2-large vs Qwen3-0.6B](https://savrn.com/models/compare/gpt2-large-vs-qwen3-0-6b)
- [OpenELM-1_1B-Instruct vs Qwen3-0.6B](https://savrn.com/models/compare/openelm-1-1b-instruct-vs-qwen3-0-6b)

## Questions About Qwen3-0.6B

### How much GPU memory does Qwen3-0.6B need?

About 1.8 GB at 16-bit and 0.5 GB at 4-bit: the weights (752M parameters) plus a working margin. A long context needs more.

### What is the cheapest GPU to run Qwen3-0.6B on?

At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.

### Can I use Qwen3-0.6B commercially?

Yes. Qwen3-0.6B is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.

### What is Qwen3-0.6B's context length?

40,960 tokens, from the maximum position embeddings in its published configuration.

## Similar Models

Model · Text generation

### [Qwen3-0.6B-Distilled-30B-A3B-Thinking-SFT](https://savrn.com/models/qwen3-0-6b-distilled-30b-a3b-thinking-sft)

[Convergent Intelligence](https://savrn.com/model-publishers/reaperdoesntknow)

A 0.6B parameter model built in two stages: knowledge distillation from a 30B Thinking teacher to establish a structured reasoning backbone, then supervised fine-tuning on legal instruction data. 50x compression. Under 500MB quantized. Runs on a phone. The training order is the thesis: teach the model how to reason first (distillation from Thinking teacher), then teach it what to reason about (legal SFT). The Thinking teacher's extended deliberation traces transfer deeper reasoning structure than an Instruct teacher — critical when the student has only 0.6B parameters to work with. Qwen3-0.6B distilled from Qwen3-30B-A3B-Thinking-2507 — a Mixture-of-Experts model with 30B total parameters…

Open weights apache-2.0 752M parameters 40,960 tokens transformers

[View model](https://savrn.com/models/qwen3-0-6b-distilled-30b-a3b-thinking-sft)

Model · Text generation

### [Qwen3-0.6B-Distilled-30B-A3B](https://savrn.com/models/qwen3-0-6b-distilled-30b-a3b)

[Convergent Intelligence](https://savrn.com/model-publishers/reaperdoesntknow)

A 0.6B parameter model distilled from Qwen3-30B-A3B-Thinking on 6,122 STEM chain-of-thought samples. 50x parameter compression. The Thinking variant teacher produces richer extended reasoning traces than the Instruct variant, transferring deeper deliberation structure into the smallest possible student. The result: a model under 500MB quantized that produces structured STEM derivations because a 30B thinking model showed it how to reason. Two key differences from standard small-model distillation: 1. Thinking teacher, not Instruct teacher. The Qwen3-30B-A3B-Thinking variant generates extended internal reasoning before committing to an answer. Its softmax distributions are higher-entropy…

Open weights apache-2.0 752M parameters 40,960 tokens transformers

[View model](https://savrn.com/models/qwen3-0-6b-distilled-30b-a3b)

Model · Text generation

### [stream-1-mini](https://savrn.com/models/stream-1-mini)

[Whisperstream](https://savrn.com/model-publishers/whisperstream)

Stream-1 Mini turns English dictation transcripts into clean Chat and Email text on your own device. It is trained to remove fillers and accidental repeats, add punctuation, and resolve spoken corrections while preserving your meaning and voice. Email mode also formats dictated greetings, paragraphs and sign-offs. This revision adds optional custom-vocabulary hints while retaining structural dictation and numbered-list support. Use these prompts and weights together. Candidates are spelling hints, not forced substitutions. Known cleanup and entity errors remain; this revision does not establish zero regressions. Pass in text from your speech recognizer. Stream-1 Mini returns the cleaned…

Open weights other 752M parameters 32,768 tokens

[View model](https://savrn.com/models/stream-1-mini)

Model · Text generation

### [Qwen3-Coder-Next-CQ-4bit](https://savrn.com/models/qwen3-coder-next-cq-4bit)

[ThousandML](https://savrn.com/model-publishers/thousandml)

Qwen3-Coder-Next quantized to 4-bit with Coded Quantization (CQ). CQ is a calibration-free 4-bit quantization method designed to preserve the behavior of the original model without requiring a calibration dataset. Because CQ does not optimize the quantization around a fixed calibration corpus, it is designed to generalize beyond calibration-specific data and workloads. This repository contains Qwen3-Coder-Next quantized with Coded Quantization (CQ) 4-bit. Calibration-free. CQ does not require a representative calibration dataset before quantization. This avoids tying the quantization procedure to a particular calibration corpus and is intended to provide better generalization beyond the…

Open weights apache-2.0 752M parameters 262,144 tokens

[View model](https://savrn.com/models/qwen3-coder-next-cq-4bit)

Model · Text generation

### [dQwen3.5-0.8B-Base](https://savrn.com/models/dqwen3-5-0-8b-base)

[IFML](https://savrn.com/model-publishers/ut-ifml)

A masked diffusion language model adapted from Qwen3.5-0.8B. The backbone is hybrid: only its attention layers are made bidirectional, and the Gated DeltaNet layers stay causal. This is a base model, with no instruction tuning. Paper: dQwen3.5: Hybrid-Attention Diffusion Language Models. Code: https://github.com/AntonXue/dQwen Needs a CUDA GPU and transformers>=5.13 (tested with torch 2.7.1+cu128, flash-linear-attention 0.5.1). generate decodes the whole canvas at once, committing positions above a confidence threshold (tau=0.9); pass blocklength=32 for left-to-right block decoding, or tau=None, stepsperblock=k for a fixed budget. The 50B-token checkpoint from the paper is…

Open weights apache-2.0 752M parameters 262,144 tokens transformers

[View model](https://savrn.com/models/dqwen3-5-0-8b-base)

Model · Text generation

### [szl-receiptagent-qwen35-0.8b-v3](https://savrn.com/models/szl-receiptagent-qwen35-0-8b-v3)

[SZL Holdings](https://savrn.com/model-publishers/szlholdings)

This is a separate research bf16 LoRA adapter (r=16, alpha=32) on Qwen/Qwen3.5-0.8B, trained with Unsloth on a laptop Blackwell GPU, with a derived merged artifact also present in the repository. "Placeholder" describes release status, not absence of weights. It does not overwrite ReceiptAgent v2 Sizes were read from the Hub listing; no weight file was downloaded or hashed by this pass, and the revision SHA was not captured. The Hub-hosted banner referenced by the szl-forge card (holo-banner.svg) is not on this tree, so it is not embedded here. The training receipt records 180 training rows and train loss 0.0537. Training loss is not evaluation. The receipt describes the local laptop…

Open weights apache-2.0 752M parameters 262,144 tokens peft

[View model](https://savrn.com/models/szl-receiptagent-qwen35-0-8b-v3)

## Qwen

[All models and datasets](https://savrn.com/model-publishers/qwen)

## Versions

- [c1899de289a0](https://savrn.com/models/qwen3-0-6b/versions/c1899de289a0) · current 2026-10-23

## Explore More

- [All text generation models](https://savrn.com/models/tasks/text-generation)
- [All models under apache-2.0](https://savrn.com/models/licenses/apache-2-0)
- [Model comparisons](https://savrn.com/models/comparisons)
- [The model directory](https://savrn.com/models)
- [Open model prices by host](https://savrn.com/ai-index/pricing/open-models)

## Source

- Repository metadata, read 2026-10-19.
- [Qwen website](https://qwen.ai/)
- [Hugging Face record](https://huggingface.co/Qwen/Qwen3-0.6B)
- [How the hub is built](https://savrn.com/model-hub/methodology)