What are the hardware requirements for GigaChat3 10B A1.8B Base?

GigaChat3 10B A1.8B Base requires approximately 22.5 GB of VRAM and supports a context window of 256K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed GigaChat3 10B A1.8B Base and how large is it?

GigaChat3 10B A1.8B Base is developed by ai-sage, a model with 10b 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 GigaChat3 10B A1.8B Base?

GigaChat3 10B A1.8B Base is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

GigaChat3 10B A1.8B Base by ai-sage — VRAM 22.5GB, 256K context

Name: GigaChat3 10B A1.8B Base
Author: ai-sage

GigaChat3 10B A1.8B Base is an open-source language model by ai-sage. Features: 10b LLM, VRAM: 22.5GB, Context: 256K, License: mit, LLM Explorer Score: 0.22.

Deepseek v3 En Moe Region:us Ru Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/ai-sage/GigaChat3-10B-A1.8B-base

GigaChat3 10B A1.8B Base Parameters and Internals

LLM Name	GigaChat3 10B A1.8B Base
Repository 🤗	https://huggingface.co/ai-sage/GigaChat3-10B-A1.8B-base
Model Size	10b
Required VRAM	22.5 GB
Updated	2026-07-15
Maintainer	ai-sage
Model Type	deepseek_v3
Model Files	2.5 GB: 0-of-13 1.6 GB: 1-of-13 1.6 GB: 2-of-13 1.6 GB: 3-of-13 1.6 GB: 4-of-13 1.6 GB: 5-of-13 1.6 GB: 6-of-13 1.6 GB: 7-of-13 1.6 GB: 8-of-13 1.6 GB: 9-of-13 1.6 GB: 10-of-13 1.6 GB: 11-of-13 1.6 GB: 12-of-13 0.8 GB: 13-of-13
Supported Languages	ru en
Model Architecture	DeepseekV3ForCausalLM
License	mit
Context Length	262144
Model Max Length	262144
Transformers Version	4.53.2
Tokenizer Class	PreTrainedTokenizer
Vocabulary Size	128256
Torch Data Type	bfloat16

Rank the GigaChat3 10B A1.8B Base 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

Best Alternatives to GigaChat3 10B A1.8B Base

Best Alternatives	Context / RAM	Downloads	Likes
GigaChat3.1 10B A1.8B	256K / 12.3 GB	5673	29

Note: green Score (e.g. "73.2") means that the model is better than ai-sage/GigaChat3-10B-A1.8B-base.