alt.freeweights.text / Holo4-35B-A3B
From: Hcompany <huggingface.co/Hcompany>Newsgroups: alt.freeweights.textSubject:

Holo4-35B-A3B

Date: 2026-11-15Commit: 5084587Signed: 2DC28CC7B26A283B ✓

Image text to text model with 35.1B parameters. Safetensors weights, 70.2 GB in all. Published by Hcompany under Apache-2.0.

35.1B params70.2GBHolo4-35B-A3Bsafetensorspermissive
FactsParams35.1BSize70.2 GBFormatssafetensorsTaskimage text to textLicenseApache-2.0From sourcepermissive
$ fw get Hcompany/Holo4-35B-A3B
# manifest sha256 7d44001a294f… · 70.2 GB · 39 files
$ fw verify ./Holo4-35B-A3B Hcompany/Holo4-35B-A3B
From sourceHosted (signed)Hcompany/Holo4-35B-A3BLast check: 2026-11-15 03:37 UTC

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

Fine-tune of Qwen3.6-35B-A3B.

Works with

Overview

Size
70.2 GB
Weights
70.2 GB (safetensors)
Parameters
35.1B
Formats
safetensors
Task
image text to text
License
Apache-2.0 (permissive)
Files
39, each with a sha256
Publisher
Hcompany
Added
2026-11-15
Section
alt.freeweights.text
Lineage

Original release from the publisher.

Base model: Qwen/Qwen3.6-35B-A3B

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
BadgeHostLast checkResult
From sourcehuggingface.co2026-11-15 03:37 UTCpassed

Model card

From the publisher at commit 5084587. Shown as plain text with links; images open at the publisher.

Holo4-35B-A3B#

Important. Holo4 family:

H Models API Agent trajectories

Model summary#

Holo4-35B-A3B is a vision-language model (VLM) for Computer Use, built on Qwen3.6-35B-A3B and developed by H Company. Used with the hai-agents harness, it can send screenshots and tool results to the model, then execute its requested clicks, typing, code, and tool calls.

| Specification | Value

| Model ID | Hcompany/Holo4-35B-A3B

| Architecture | Qwen3.6 MoE

| Checkpoint format | BF16 safetensors

| Maximum context length in config | 262,144 tokens

This demo shows Holo4-27B using FreeCAD to build a replica of the Eiffel Tower. More examples in the blog post.

Prompt

Build a 3D Eiffel Tower in FreeCAD at a scale of 1 mm per metre. Center it on the origin and align it with the X and Y axes. Its plan must stay square at every height. The distance from the center to each corner is 62.5 mm at ground level, 32.5 mm at height 57, 17.5 mm at height 115, and 9.35 mm at height 276. Connect these widths with a smooth curve that narrows quickly near the base and more slowly near the top.

Make four separate, identical square legs, one in each quadrant. Their outer corners follow that curve, and each leg narrows from 14 mm across at ground level to 4 mm at height 276. Leave the space between the legs open. Add centered square platforms measuring 72 mm by 72 mm by 4 mm at height 57, 40 mm by 40 mm by 3 mm at height 115, and 22 mm by 22 mm by 3 mm at height 276. Add a square mast from height 276 to 324, tapering from 8 mm across to 2 mm across. Make every component a closed solid with nonzero volume, without filling the space between the legs.

Usage#

The harness sends screenshots and tool results to Holo4, executes the model's requested actions, and sends the results back. It can give the model access to application tools and code execution.

Refer to the documentation for more details about:

Performance#

Holo4 models improve significantly over their Qwen base models. On OSWorld, Holo4-27B scores 85.2% at $0.08 per task. We also evaluate them on Agentic Task Factory, a set of held-out business workflows across web, desktop, and MCP tools.

Benchmark results#

Image: Holo4 benchmark comparison table

Read the blog post for more tables about the benchmark results.

Pareto plots#

These plots compare benchmark scores with cost per task. On OSWorld 2.0, Holo4-27B scores 61.7% at $1.22 per task, and Holo4-35B-A3B scores 30.9% at $0.61 per task.

Image: Cost versus score on OSWorld 2.0

On AutomationBench, Holo4-27B scores 45.4% at $0.05 per task, and Holo4-35B-A3B scores 34.5% at $0.02 per task.

Image: Cost versus score on AutomationBench

Open-source evaluation traces#

For transparency, we share all agent trajectories in the open-source dataset at Hcompany/trajectories.

Training#

Image: Holo4 training pipeline

License#

The model weights are available under the Apache License 2.0. This model is built on Qwen3.6-35B-A3B, which Alibaba Cloud also releases under the Apache License 2.0.

Files

39 files · 70.2 GBWeights 70.2 GB (safetensors). Sizes and sha256 from the signed manifest.
Show all 39 files with their sha256
FileSizesha256
.gitattributes1.57 KB34448b82c17d60fec9b65b1f093c115ddbaadc04beb1b0140b6bfed2e012a930
LICENSE11.3 KB50cbab8a892c5f2993b8c7351a99182507472def3b1374558308605d99b86b32
README.md6.08 KB4f1aa5006aa9adaeb5ac549e5a5f8ea164352714a00528b1afb8602812d45b44
chat_template.jinja7.76 KBe84f32a23fdda27689f868aa4a1a5621f41133e51a48d7f3efcbea2839574259
config.json3.19 KB6fa9fe6c73d79a30dfd40809d43ac8cd96ad74966d2329716c8399c785876677
generation_config.json213 Ba219dff6ed1d52d8a88fe50026de36ebcbcb913cbac0e70fdfaecf380949d7f1
merges.txt3.35 MBa9d356d7bdf1ef4949e3e748e95b8e10ad9d4e2e838eddc38a0a7b6b94d1db8d
model-00001-of-00026.safetensors4.00 GB3b391b0877531f542dddd3ed5dc01657f205884ccdb2d6629617d12fb0f35c13
model-00002-of-00026.safetensors1.28 GB64da597fb5181c1c5ee2a884d8477eb887d98f01c9d7e67568b70f6aa7bc8214
model-00003-of-00026.safetensors3.36 GB077681a9e3ebc9056fa8dad078b1e81f4b4256a555eea061b19e1d3fdf71e304
model-00004-of-00026.safetensors3.37 GB93573262673c3890a11f7bdc9f2918ba4b02fe1a9a730e61b39db6d579054fc0
model-00005-of-00026.safetensors3.36 GBeb877d57a4b5d7bc235d597faccb5ee77c8a212dcddc9a660183fb4676c30001
model-00006-of-00026.safetensors3.96 GB412382c3e5002a1b420a1fc238a659b0576318f8d652f4ccc8b6fc3a1140427e
model-00007-of-00026.safetensors1.10 GB9464346998f119de75252c2330af2a6c3d560e26f7c4e5d73d4b9466da95e333
model-00008-of-00026.safetensors3.95 GBfc077a6535849393fd02e4f8740c1ed7778ad8fcb57e9cf4fe284f3185ca9903
model-00009-of-00026.safetensors1.10 GBf7b2610c96827978aae2a48fe527562ba4d9797b4eb32fe1a222659a21d4a2b4
model-00010-of-00026.safetensors3.95 GB5da556950e1d7d9f3fc482e1bddfa182ae1aece916bfaf96ec2653255863c2d3
model-00011-of-00026.safetensors1.10 GBd0d41d7c8804e247db07fcd8735084034de48d2158480b23396ce0cebd01968c
model-00012-of-00026.safetensors3.41 GBd9a75c85a68156b00a25f62ebef5a6d188009caf4588a2f980af3227ff65b249
model-00013-of-00026.safetensors1.63 GBb293a501475ab68ae315a5fc2ad91d18a587b8b85cca65cdc22fbd8d3db36c4a
model-00014-of-00026.safetensors3.42 GB75dd03dadbe20dd92be3fa586139c1184f12c7c3e498b95522229cc14ed3f433
model-00015-of-00026.safetensors1.63 GB8443d8d6dc42a6dfe773d18a387b5ed5fcba91f8b8f7b29be7072e0c80b8a254
model-00016-of-00026.safetensors3.95 GB88ee83c76b89c8971564605d3a7762a25d5d20be9d3cafef9fd185359d5f25b3
model-00017-of-00026.safetensors1.10 GB0165a48a2de0680c9c79275de0a166add00a9dd58c5d61f0374597477441db2b
model-00018-of-00026.safetensors3.95 GB15b882af9483db2ec3c5c45beb57d907e3d923b3f7cfef3fd838cc5e536baf9e
model-00019-of-00026.safetensors1.10 GB466ce53766c07af0c81a4dbd1ceebbc03d1c7d724c55602ad68bdbfc5bfafb13
model-00020-of-00026.safetensors3.41 GB9787aaa36640979c4512096be07c8bdd747ecd5a800b7d2170ea112c30ae663a
model-00021-of-00026.safetensors1.63 GB16f6f945357a30a4ff0e9ccbf12924159cc237ccf1206ec3f8d59c67c59ce14f
model-00022-of-00026.safetensors3.37 GB4c680ec96644810500236e9ec03988b8a979531c1d15af3f954eb0abec4e7421
model-00023-of-00026.safetensors3.36 GB6ea2f065a4aeafe4c14fcdaf6356297d00dee34782a3a43e19f2e5cae077c1b1
model-00024-of-00026.safetensors3.37 GBb3123fe991897aa9aa1b8e94154f80d885834d72302e69c2f25de756b9345322
model-00025-of-00026.safetensors2.76 GBd5240d120d9aa501f14b7a889fb6032545d983314b9eef636907b772c32247ef
model-00026-of-00026.safetensors1.62 GB1bd86d77b6847bb29d13249fc30d066a3753e36b2d8056e2bfe97edc89476357
model.safetensors.index.json96.9 KBf011e20fec67da1c980230961f8af26555aa2fb11159d5f651bee90552ef605c
preprocessor_config.json390 B27225450ac9c6529872ee1924fcb0962ff5634834f817040f444118116f4e516
tokenizer.json12.8 MB5f9e4d4901a92b997e463c1f46055088b6cca5ca61a6522d1b9f64c4bb81cb42
tokenizer_config.json16.7 KB5186f0defcd7f232382c7f0aebcd2252d073bb921ab240e407b7ae8745d2b29b
video_preprocessor_config.json385 B7768af27c1fafa9cc9011c1dc20067e03f8915e03b63504550e11d5066986d13
vocab.json6.72 MBce99b4cb2983d118806ce0a8b777a35b093e2000a503ebde25853284c9dfa003
VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commitDateContents
5084587278 pinned now2026-11-15 03:37 UTC39 files · 70.2 GB

Use it

$ fw get Hcompany/Holo4-35B-A3B
# manifest sha256 7d44001a294f… · 70.2 GB · 39 files
$ fw verify ./Holo4-35B-A3B Hcompany/Holo4-35B-A3B
fw.pyChecks the signature and every sha256, tries each source in order, resumes.
$ python3 fw.py get https://freeweights.allternit.com/manifests/Hcompany/Holo4-35B-A3B.json --dir ./Holo4-35B-A3B $ python3 fw.py verify ./Holo4-35B-A3B ./Hcompany__Holo4-35B-A3B.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download Hcompany/Holo4-35B-A3B --revision 508458727831368da85d53a8055c2b0f923ed4f3 --local-dir ./Holo4-35B-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("Hcompany/Holo4-35B-A3B", revision="508458727831368da85d53a8055c2b0f923ed4f3") model = AutoModelForImageTextToText.from_pretrained("Hcompany/Holo4-35B-A3B", revision="508458727831368da85d53a8055c2b0f923ed4f3", device_map="auto")
vLLMServe an OpenAI-compatible API.
$ HF_ENDPOINT=https://freeweights.allternit.com vllm serve Hcompany/Holo4-35B-A3B --revision 508458727831368da85d53a8055c2b0f923ed4f3

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. 2026-11-15
Text in
LICENSE
Text sha256
50cbab8a892c5f2993b8c7351a99182507472def3b1374558308605d99b86b32
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: Hcompany <huggingface.co/Hcompany>Newsgroups: alt.freeweights.textSubject: Holo4-35B-A3BDate: 2026-11-15Commit: 5084587Signed: 2DC28CC7B26A283B ✓
Upstream
Hcompany/Holo4-35B-A3B
Pinned commit
508458727831368da85d53a8055c2b0f923ed4f3
Manifest
Holo4-35B-A3B.json · .minisig
Manifest sha256
7d44001a294f7531fe2d616ab25a8991d847c005db538373a4723cdbfce2ed0a
Signature
minisign, Ed25519, key 2DC28CC7B26A283B
Created
2026-11-15 03:37 UTC
Generator
freeweights-manifest/0.1.0
Takedowns
None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/Hcompany/Holo4-35B-A3B.json $ curl -O https://freeweights.allternit.com/manifests/Hcompany/Holo4-35B-A3B.json.minisig $ minisign -Vm Holo4-35B-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

Model tree

Parent

Qwen/Qwen3.6-35B-A3B · this model is a finetune of it

1 model in the catalog name this one as their parent.

Quantizations
1
Get it
Download
$ HF_ENDPOINT=https://freeweights.allternit.com hf download Hcompany/Holo4-35B-A3B
Open in Freeweights Studio

Pull, verify and run it in the Allternit app on Desktop, web or phone.

Signed manifest
minisign, key 2DC28CC7B26A283B. Holo4-35B-A3B.json. Provenance. Takedown history: none.

Variants

1 quantized or adapter build of Holo4-35B-A3B. Each has its own license check and page.

From: laion <huggingface.co/laion>Subject: a2-rl-e2egit_large-50-32B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_exp_rpt_e2egit-large-15-8B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_exp_rpt_e2egit-v2-10-8B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_exp_rpt_pymethods2test-large-80-8B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_exp_rpt_unitsyn-python-large-10-8B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_mix_h2_language_proportional-65-8B