What are the hardware requirements for Thomson 1.0 Small?

Thomson 1.0 Small requires approximately 70.5 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Thomson 1.0 Small and how large is it?

Thomson 1.0 Small is developed by thomsonreuters, a model with 35.1b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Thomson 1.0 Small?

Thomson 1.0 Small is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Thomson 1.0 Small by thomsonreuters — VRAM 70.5GB

Name: Thomson 1.0 Small
Author: thomsonreuters

Thomson 1.0 Small is an open-source language model by thomsonreuters. Features: 35.1b LLM, VRAM: 70.5GB, License: other, LLM Explorer Score: 0.31.

Base model:finetune:tri-fair-l... Base model:tri-fair-lab/snowdo... Conversational Endpoints compatible Image-text-to-text Qwen3 5 moe Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/thomsonreuters/Thomson-1.0-Small

Thomson 1.0 Small Parameters and Internals

LLM Name	Thomson 1.0 Small
Repository 🤗	https://huggingface.co/thomsonreuters/Thomson-1.0-Small
Base Model(s)	tri-fair-lab/Snowdon1.1-Small tri-fair-lab/Snowdon1.1-Small
Model Size	35.1b
Required VRAM	70.5 GB
Updated	2026-08-26
Maintainer	thomsonreuters
Model Type	qwen3_5_moe
Model Files	5.0 GB: 1-of-16 4.1 GB: 2-of-16 4.5 GB: 3-of-16 4.1 GB: 4-of-16 4.5 GB: 5-of-16 4.1 GB: 6-of-16 4.5 GB: 7-of-16 4.1 GB: 8-of-16 4.5 GB: 9-of-16 4.1 GB: 10-of-16 4.5 GB: 11-of-16 4.0 GB: 12-of-16 5.0 GB: 13-of-16 4.5 GB: 14-of-16 4.0 GB: 15-of-16 5.0 GB: 16-of-16
Model Architecture	Qwen3_5MoeForConditionalGeneration
License	other
Model Max Length	262144
Transformers Version	5.3.0
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Errors	replace

Rank the Thomson 1.0 Small capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Quantized Models of the Thomson 1.0 Small

Model	Likes	Downloads	VRAM
Thomson 1.0 Small AWQ INT4	2	17	24 GB
Thomson 1.0 Small Mlx 8Bit	2	23	37 GB

Best Alternatives to Thomson 1.0 Small

Best Alternatives	Context / RAM	Downloads	Likes
Nex N2.5 Mini	0K / 70.4 GB	3121	685
Agents A1	0K / 70.6 GB	23272	629
Nex N2 Mini	0K / 70.4 GB	2289	294
Huihui Agents A1 Abliterated	0K / 70.6 GB	3339	29
Occamy 1.0 FP8	0K / 36.8 GB	434	4
Nex N2 Mini	0K / 70.4 GB	219	1
...yJ5ua3Sebfyr8kcVwLAdaiJLUC8pkW	0K / 70.4 GB	1894	0
Affine 5g4yy75zuz T6	0K / 70.2 GB	756	0
Affine 5efg6cm3yl King	0K / 70.3 GB	578	0
Affine 5ekxlcg3fx Abc	0K / 70.3 GB	863	0

Note: green Score (e.g. "73.2") means that the model is better than thomsonreuters/Thomson-1.0-Small.