alt.freeweights.qwen / Qwen3.5-35B-A3B

From:

Qwen <huggingface.co/Qwen>Newsgroups: alt.freeweights.qwenSubject:

Qwen3.5-35B-A3B

Date: 2026-11-14Commit: 59d61f3Signed: 2DC28CC7B26A283B ✓

Image text to text model with 36B parameters. Safetensors weights, 71.9 GB in all. Published by Qwen under Apache-2.0.

36B params

71.9GBQwen3.5-35B-A3Bsafetensorspermissive

FactsParams36BSize71.9 GBFormatssafetensorsTaskimage text to textLicenseApache-2.0From sourcepermissive

$ fw get Qwen/Qwen3.5-35B-A3B
# manifest sha256 dc2b303fc5f9… · 71.9 GB · 27 files
$ fw verify ./Qwen3.5-35B-A3B Qwen/Qwen3.5-35B-A3B

From sourceHosted (signed)Qwen/Qwen3.5-35B-A3BLast check: 2026-11-14 21:12 UTC

Run at home Get help

Get help: custom fine-tune or private deploy. We review your setup before we quote.

Fine-tune of Qwen3.5-35B-A3B-Base.

Works with

Run23 Fine-tune29 Convert15 Evaluate13 Scan13 Apps62

Runs with

Ranked by fit to this model (safetensors), then by GitHub stars.

vLLM exo Text Generation WebUI

SGLang MLC LLM KTransformers LMCache Xinference LMDeploy mistral.rs Mooncake FlashInfer

Show all 23

AIBrix fastllm ExLlamaV3

Transformers LocalAI tinygrad Triton Inference Server BentoML CTranslate2 ExLlamaV2 LitServe

Convert with

mergekit TensorRT Model Optimizer

LLM Compressor Neural Compressor AutoRound AQLM GPTQModel ONNX Core ML Tools ExecuTorch

Show all 15

Evaluate with

Opik DeepEval Ragas lm-evaluation-harness OpenCompass SWE-bench

simple-evals TruLens HELM

LightEval

Evaluate Weave

Show all 13

Autoevals

Scan with

promptfoo garak Guardrails NeMo Guardrails Giskard DeepTeam AI Fairness 360 Fairlearn Detoxify MCP Scanner AgentDojo ModelScan

Show all 13

picklescan

Overview Model card Files Use it License Provenance Model tree

Overview

Size: 71.9 GB
Weights: 71.9 GB (safetensors)
Parameters: 36B
Formats: safetensors
Task: image text to text
License: Apache-2.0 (permissive)
Files: 27, each with a sha256
Publisher: Qwen
Added: 2026-11-14
Section: alt.freeweights.qwen

Lineage

Original release from the publisher.

Base model: Qwen/Qwen3.5-35B-A3B-Base

Where these bytes come from

Publisher at the pinned commit→You

Sources are tried in order. A file counts only when its size and sha256 match the manifest. Sources are checked when you ask for them, not on a timer.

Sources

Badge	Host	Last check	Result
From source	huggingface.co	2026-11-14 21:12 UTC	passed

Model card

From the publisher at commit 59d61f3. Shown as plain text with links; images open at the publisher.

Qwen3.5-35B-A3B#

Image: image

Qwen Chat

Note. This repository contains model weights and configuration files for the post-trained model in the Hugging Face Transformers format.

These artifacts are compatible with Hugging Face Transformers, vLLM, SGLang, KTransformers, etc.

Tip. For users seeking managed, scalable inference without infrastructure maintenance, the official Qwen API service is provided by Alibaba Cloud Model Studio.

In particular, Qwen3.5-Flash is the hosted version corresponding to Qwen3.5-35B-A3B with more production features, e.g., 1M context length by default and official built-in tools. For more information, please refer to the User Guide.

Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance. Qwen3.5 represents a significant leap forward, integrating breakthroughs in multimodal learning, architectural efficiency, reinforcement learning scale, and global accessibility to empower developers and enterprises with unprecedented capability and efficiency.

Qwen3.5 Highlights#

Qwen3.5 features the following enhancement:

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.
Scalable RL Generalization: Reinforcement learning scaled across million-agent environments with progressively complex task distributions for robust real-world adaptability.
Global Linguistic Coverage: Expanded support to 201 languages and dialects, enabling inclusive, worldwide deployment with nuanced cultural and regional understanding.
Next-Generation Training Infrastructure: Near-100% multimodal training efficiency compared to text-only training and asynchronous RL frameworks supporting massive-scale agent scaffolds and environment orchestration.

Image: Benchmark Results

For more details, please refer to our blog post Qwen3.5.

Model Overview#

Type: Causal Language Model with Vision Encoder
Training Stage: Pre-training & Post-training
Language Model
- Number of Parameters: 35B in total and 3B activated
- Hidden Dimension: 2048
- Token Embedding: 248320 (Padded)
- Number of Layers: 40
- Hidden Layout: 10 × (3 × (Gated DeltaNet → MoE) → 1 × (Gated Attention → MoE))
- Gated DeltaNet:
  - Number of Linear Attention Heads: 32 for V and 16 for QK
  - Head Dimension: 128
- Gated Attention:
  - Number of Attention Heads: 16 for Q and 2 for KV
  - Head Dimension: 256
  - Rotary Position Embedding Dimension: 64
- Mixture Of Experts
  - Number of Experts: 256
  - Number of Activated Experts: 8 Routed + 1 Shared
  - Expert Intermediate Dimension: 512
- LM Output: 248320 (Padded)
- MTP: trained with multi-steps
Context Length: 262,144 natively and extensible up to 1,010,000 tokens.

Benchmark Results#

Language#

| Knowledge

| MMLU-Pro | 83.7 | 80.8 | 84.4 | 86.7 | 86.1 | 85.3

| MMLU-Redux | 93.7 | 91.0 | 93.8 | 94.0 | 93.2 | 93.3

| C-Eval | 82.2 | 76.2 | 92.1 | 91.9 | 90.5 | 90.2

| SuperGPQA | 58.6 | 54.6 | 64.9 | 67.1 | 65.6 | 63.4

| Instruction Following

| IFEval | 93.9 | 88.9 | 87.8 | 93.4 | 95.0 | 91.9

| IFBench | 75.4 | 69.0 | 51.7 | 76.1 | 76.5 | 70.2

| MultiChallenge | 59.0 | 45.3 | 50.2 | 61.5 | 60.8 | 60.0

| Long Context

| AA-LCR | 68.0 | 50.7 | 60.0 | 66.9 | 66.1 | 58.5

| LongBench v2 | 56.8 | 48.2 | 54.8 | 60.2 | 60.6 | 59.0

| STEM & Reasoning

| HLE w/ CoT | 19.4 | 14.9 | 18.2 | 25.3 | 24.3 | 22.4

| GPQA Diamond | 82.8 | 80.1 | 81.1 | 86.6 | 85.5 | 84.2

| HMMT Feb 25 | 89.2 | 90.0 | 85.1 | 91.4 | 92.0 | 89.0

| HMMT Nov 25 | 84.2 | 90.0 | 89.5 | 90.3 | 89.8 | 89.2

| Coding

| SWE-bench Verified | 72.0 | 62.0 | -- | 72.0 | 72.4 | 69.2

| Terminal Bench 2 | 31.9 | 18.7 | -- | 49.4 | 41.6 | 40.5

| LiveCodeBench v6 | 80.5 | 82.7 | 75.1 | 78.9 | 80.7 | 74.6

| CodeForces | 2160 | 2157 | 2146 | 2100 | 1899 | 2028

| OJBench | 40.4 | 41.5 | 32.7 | 39.5 | 40.1 | 36.0

| FullStackBench en | 30.6 | 58.9 | 61.1 | 62.6 | 60.1 | 58.1

| FullStackBench zh | 35.2 | 60.4 | 63.1 | 58.7 | 57.4 | 55.0

| General Agent

| BFCL-V4 | 55.5 | -- | 54.8 | 72.2 | 68.5 | 67.3

| TAU2-Bench | 69.8 | -- | 58.5 | 79.5 | 79.0 | 81.2

| VITA-Bench | 13.9 | -- | 31.6 | 33.6 | 41.9 | 31.9

| DeepPlanning | 17.9 | -- | 17.1 | 24.1 | 22.6 | 22.8

| Search Agent

| HLE w/ tool | 35.8 | 19.0 | -- | 47.5 | 48.5 | 47.4

| Browsecomp | 48.1 | 41.1 | -- | 63.8 | 61.0 | 61.0

| Browsecomp-zh | 49.5 | 42.9 | -- | 69.9 | 62.1 | 69.5

| WideSearch | 47.2 | 40.4 | -- | 60.5 | 61.1 | 57.1

| Seal-0 | 34.2 | 45.1 | -- | 44.1 | 47.2 | 41.4

| Multilingualism

| MMMLU | 86.2 | 78.2 | 83.4 | 86.7 | 85.9 | 85.2

| MMLU-ProX | 78.5 | 74.5 | 77.9 | 82.2 | 82.2 | 81.0

| NOVA-63 | 51.9 | 51.1 | 55.4 | 58.6 | 58.1 | 57.1

| INCLUDE | 81.8 | 74.0 | 81.0 | 82.8 | 81.6 | 79.7

| Global PIQA | 88.5 | 84.1 | 85.7 | 88.4 | 87.5 | 86.6

| PolyMATH | 67.3 | 54.0 | 60.1 | 68.9 | 71.2 | 64.4

| WMT24++ | 80.7 | 74.4 | 75.8 | 78.3 | 77.6 | 76.3

| MAXIFE | 85.3 | 83.7 | 83.2 | 87.9 | 88.0 | 86.6

CodeForces: evaluated on our own query set.
TAU2-Bench: we follow the official setup except for the airline domain, where all models are evaluated by applying the fixes proposed in the Claude Opus 4.5 system card.
Search Agent: most search agents built on our model adopt a simple context-folding strategy(256k): once the cumulative Tool Response length reaches a preset threshold, earlier Tool Responses are pruned from the history to keep the context within limits.
WideSearch: we use a 256k context window without any context management.
MMLU-ProX: we report the averaged accuracy on 29 languages.
WMT24++: a harder subset of WMT24 after difficulty labeling and rebalancing; we report the averaged scores on 55 languages using XCOMET-XXL.
MAXIFE: we report the accuracy on English + multilingual original prompts (totally 23 settings).
Empty cells (--) indicate scores not yet available or not applicable.

Vision Language#

| STEM and Puzzle

| MMMU | 79.0 | 79.6 | 80.6 | 83.9 | 82.3 | 81.4

| MMMU-Pro | 67.3 | 68.4 | 69.3 | 76.9 | 75.0 | 75.1

| MathVision | 71.9 | 71.1 | 74.6 | 86.2 | 86.0 | 83.9

| Mathvista(mini) | 79.1 | 79.8 | 85.8 | 87.4 | 87.8 | 86.2

| DynaMath | 81.4 | 78.8 | 82.8 | 85.9 | 87.7 | 85.0

| ZEROBench | 3 | 4 | 4 | 9 | 10 | 8

| ZEROBench_sub | 27.3 | 26.3 | 28.4 | 36.2 | 36.2 | 34.1

| VlmsAreBlind | 75.8 | 85.5 | 79.5 | 96.7 | 96.9 | 97.0

| BabyVision | 20.9 | 18.6 | 22.2 | 40.2 / 34.5 | 44.6 / 34.8 | 38.4 / 29.6

| General VQA

| RealWorldQA | 79.0 | 70.3 | 81.3 | 85.1 | 83.7 | 84.1

| MMStar | 74.1 | 73.8 | 78.7 | 82.9 | 81.0 | 81.9

| MMBenchEN-DEV-v1.1 | 86.8 | 88.3 | 89.7 | 92.8 | 92.6 | 91.5

| SimpleVQA | 56.8 | 57.6 | 61.3 | 61.7 | 56.0 | 58.3

| HallusionBench | 63.2 | 59.9 | 66.7 | 67.6 | 70.0 | 67.9

| Text Recognition and Document Understanding

| OmniDocBench1.5 | 77.0 | 85.8 | 84.5 | 89.8 | 88.9 | 89.3

| CharXiv(RQ) | 68.6 | 67.2 | 66.1 | 77.2 | 79.5 | 77.5

| MMLongBench-Doc | 50.3 | -- | 56.2 | 59.0 | 60.2 | 59.5

| CC-OCR | 70.8 | 68.1 | 81.5 | 81.8 | 81.0 | 80.7

| AI2D_TEST | 88.2 | 87.0 | 89.2 | 93.3 | 92.9 | 92.6

| OCRBench | 82.1 | 76.6 | 87.5 | 92.1 | 89.4 | 91.0

| Spatial Intelligence

| ERQA | 54.0 | 45.0 | 52.5 | 62.0 | 60.5 | 64.8

| CountBench | 91.0 | 90.0 | 93.7 | 97.0 | 97.8 | 97.8

| RefCOCO(avg) | -- | -- | 91.1 | 91.3 | 90.9 | 89.2

| ODInW13 | -- | -- | 43.2 | 44.5 | 41.1 | 42.6

| EmbSpatialBench | 80.7 | 71.8 | 84.3 | 83.9 | 84.5 | 83.1

| RefSpatialBench | 9.0 | 2.2 | 69.9 | 69.3 | 67.7 | 63.5

| LingoQA | 62.4 | 12.8 | 66.8 | 80.8 | 82.0 | 79.2

| Hypersim | -- | -- | 11.0 | 12.7 | 13.0 | 13.1

| SUNRGBD | -- | -- | 34.9 | 36.2 | 35.4 | 33.4

| Nuscene | -- | -- | 13.9 | 15.4 | 15.2 | 14.6

| Video Understanding

| VideoMME(w sub.) | 83.5 | 81.1 | 83.8 | 87.3 | 87.0 | 86.6

The publisher's README is longer; it is cut here. Read the rest at the source.

Files

27 files · 71.9 GBWeights 71.9 GB (safetensors). Sizes and sha256 from the signed manifest.

Show all 27 files with their sha256

File	Size	sha256
.gitattributes	1.57 KB	34448b82c17d60fec9b65b1f093c115ddbaadc04beb1b0140b6bfed2e012a930
LICENSE	11.5 KB	bbedc3fda3305820b977265f01b8619d87570a6739de3a5582c3464840f1e57a
README.md	93.2 KB	71c3bbc2eadadd5980bbdbb15df958c27d42ede24821015602a021109aacf492
chat_template.jinja	7.76 KB	a4aee8afcf2e0711942cf848899be66016f8d14a889ff9ede07bca099c28f715
config.json	3.54 KB	5e4d7f74fec2f360eb9cfbfcd6ec0c4c76e684d3a11caaed259d9fd9bfbc7944
generation_config.json	244 B	4f25002776b741773666203dcea8f54619f177ace3ae483d311102092a4658e0
merges.txt	3.35 MB	a9d356d7bdf1ef4949e3e748e95b8e10ad9d4e2e838eddc38a0a7b6b94d1db8d
model.safetensors-00001-of-00014.safetensors	5.37 GB	1b9f16fdae2411c0c50dbc7a5dacaebd033b28f3fd083e708f53cda3420e7247
model.safetensors-00002-of-00014.safetensors	5.37 GB	79f06bc069d273b37ae7e1ef21eb24d8641ac74227b6139b00af55addf27e9ee
model.safetensors-00003-of-00014.safetensors	5.37 GB	2cb56343806fd845eb2ca3ff5c46b1de409e05dc09d03edb86f7fb2d752b4f3d
model.safetensors-00004-of-00014.safetensors	5.37 GB	39a39c799e52db2ca3d6123dad077b7dbfff3dbad4bcc9cbf66502db7be650a3
model.safetensors-00005-of-00014.safetensors	5.37 GB	1a3b31ebb04bffa60d0ad5d9f2973d476d88959bb842ce003f068a6a9c5d4776
model.safetensors-00006-of-00014.safetensors	5.37 GB	af5f17d292df6b200ee928d7dac30a66eab3e8f6d5d88a84fe1d27859d4a64fe
model.safetensors-00007-of-00014.safetensors	5.37 GB	52885dbecd4e8e2e52e6cd50a1693a640ce622f219c35bf317aad89f72d956fc
model.safetensors-00008-of-00014.safetensors	5.37 GB	43f232ba029dc4ca5ad994e38913c6ff74f110198e005ee2f35e42f9ce438cb3
model.safetensors-00009-of-00014.safetensors	5.26 GB	8fa6a375dd539822b19ad18ad002f8acedd4920ff770bc284cfb428a213c9f17
model.safetensors-00010-of-00014.safetensors	5.37 GB	d89e80aaf84c0b8a442162fc97c38fe7ce85d11ae728f67aa449350da86f8212
model.safetensors-00011-of-00014.safetensors	5.37 GB	5b774025e5ed974256ce40498bfe4e1ad9ae8a4f13f9e9145666d24ea7404cae
model.safetensors-00012-of-00014.safetensors	5.37 GB	d0f0bd7a5f09e6e05aee9a6ef5ba97fe678b0e19b891a81969682b0bdc9bb5c7
model.safetensors-00013-of-00014.safetensors	5.37 GB	da8cc27a2a99eeba5674bfef03c80b3c4f08edfd839ad319a7702fcaedfd874f
model.safetensors-00014-of-00014.safetensors	2.22 GB	d5e08a7dd670d7ef8da38e7af29cb8cdf42ecded5dfb2d9fcaf06cae79dfaa41
model.safetensors.index.json	187 KB	d8d0b7ca4e61ae107e3e87a3ff21136b3ac7c789e64bb24267227ca804e04205
preprocessor_config.json	390 B	27225450ac9c6529872ee1924fcb0962ff5634834f817040f444118116f4e516
tokenizer.json	12.8 MB	5f9e4d4901a92b997e463c1f46055088b6cca5ca61a6522d1b9f64c4bb81cb42
tokenizer_config.json	16.7 KB	316230d6a809701f4db5ea8f8fc862bc3a6f3229c937c174e674ff3ca0a64ac8
video_preprocessor_config.json	385 B	7768af27c1fafa9cc9011c1dc20067e03f8915e03b63504550e11d5066986d13
vocab.json	6.72 MB	ce99b4cb2983d118806ce0a8b777a35b093e2000a503ebde25853284c9dfa003

VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.

Upstream commit	Date	Contents
59d61f3ce6 pinned now manifest updated, same commit	2026-11-14 21:12 UTC	27 files · 71.9 GB
59d61f3ce6	2026-11-14 21:12 UTC	27 files · 71.9 GB

Use it

$ fw get Qwen/Qwen3.5-35B-A3B
# manifest sha256 dc2b303fc5f9… · 71.9 GB · 27 files
$ fw verify ./Qwen3.5-35B-A3B Qwen/Qwen3.5-35B-A3B

fw.pyChecks the signature and every sha256, tries each source in order, resumes.

$ python3 fw.py get https://freeweights.allternit.com/manifests/Qwen/Qwen3.5-35B-A3B.json --dir ./Qwen3.5-35B-A3B
$ python3 fw.py verify ./Qwen3.5-35B-A3B ./Qwen__Qwen3.5-35B-A3B.json

hfThe Hugging Face CLI with one setting. Same repo id.

$ HF_ENDPOINT=https://freeweights.allternit.com hf download Qwen/Qwen3.5-35B-A3B --revision 59d61f3ce65a6d9863b86d2e96597125219dc754 --local-dir ./Qwen3.5-35B-A3B

transformersLoad the weights in Python.

import os
os.environ["HF_ENDPOINT"] = "https://freeweights.allternit.com"
from transformers import AutoModelForImageTextToText, AutoProcessor
processor = AutoProcessor.from_pretrained("Qwen/Qwen3.5-35B-A3B", revision="59d61f3ce65a6d9863b86d2e96597125219dc754")
model = AutoModelForImageTextToText.from_pretrained("Qwen/Qwen3.5-35B-A3B", revision="59d61f3ce65a6d9863b86d2e96597125219dc754", device_map="auto")

vLLMServe an OpenAI-compatible API.

$ HF_ENDPOINT=https://freeweights.allternit.com vllm serve Qwen/Qwen3.5-35B-A3B --revision 59d61f3ce65a6d9863b86d2e96597125219dc754

Run at home

License

Name: Apache License 2.0
Id: Apache-2.0
Class: Permissive
Redistribution: yes
Listing: Hosted (signed)
Decided: License as declared by the publisher. No person read it. 2026-11-14
Text in: LICENSE
Text sha256: bbedc3fda3305820b977265f01b8619d87570a6739de3a5582c3464840f1e57a
Terms gate: No extra terms beyond the license.

Read the license text

Licenses are checked against the license table. A custom license passes only after a person reads it.

Provenance

From:

Qwen <huggingface.co/Qwen>Newsgroups: alt.freeweights.qwenSubject: Qwen3.5-35B-A3BDate: 2026-11-14Commit: 59d61f3Signed: 2DC28CC7B26A283B ✓

Upstream: Qwen/Qwen3.5-35B-A3B
Pinned commit: 59d61f3ce65a6d9863b86d2e96597125219dc754
Manifest: Qwen3.5-35B-A3B.json · .minisig
Manifest sha256: dc2b303fc5f97227c9ecbd7a4d5b7008a19d042b02bba1e83f29f79e22e7a05b
Signature: minisign, Ed25519, key 2DC28CC7B26A283B
Created: 2026-11-14 21:12 UTC
Generator: freeweights-manifest/0.1.0
Takedowns: None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/Qwen/Qwen3.5-35B-A3B.json
$ curl -O https://freeweights.allternit.com/manifests/Qwen/Qwen3.5-35B-A3B.json.minisig
$ minisign -Vm Qwen3.5-35B-A3B.json -P RWQ7KGqyx4zCLXtUFOqvfv8L3g+8fpKJ1FsuxPyOYMF+4KufbUxKq1Oq
$ shasum -a 256 <file>   # compare with the Files tab

A manifest whose signature fails must be rejected. A file whose size or sha256 does not match must be discarded. Verify a download