What are the hardware requirements for GigaChat3.5 432B A28B Base?

GigaChat3.5 432B A28B Base requires approximately 724.5 GB of VRAM and supports a context window of 128K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed GigaChat3.5 432B A28B Base and how large is it?

GigaChat3.5 432B A28B Base is developed by ai-sage, a model with 432b 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 GigaChat3.5 432B A28B Base?

GigaChat3.5 432B A28B Base is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

GigaChat3.5 432B A28B Base by ai-sage — VRAM 724.5GB, 128K context

Name: GigaChat3.5 432B A28B Base
Author: ai-sage

GigaChat3.5 432B A28B Base is an open-source language model by ai-sage. Features: 432b LLM, VRAM: 724.5GB, Context: 128K, License: mit, LLM Explorer Score: 0.27.

Custom code En Gigachat3 5 Hybrid-attention Instruct Linear-attention Long-context Moe Mtp Multilingual Region:us Ru Safetensors Sharded Tensorflow Tool-use

Model Card on HF 🤗: https://huggingface.co/ai-sage/GigaChat3.5-432B-A28B-base

GigaChat3.5 432B A28B Base Parameters and Internals

LLM Name	GigaChat3.5 432B A28B Base
Repository 🤗	https://huggingface.co/ai-sage/GigaChat3.5-432B-A28B-base
Model Size	432b
Required VRAM	724.5 GB
Updated	2026-08-07
Maintainer	ai-sage
Model Type	gigachat3_5
Model Files	16.1 GB: 1-of-54 16.1 GB: 2-of-54 16.1 GB: 3-of-54 16.1 GB: 4-of-54 16.1 GB: 5-of-54 16.1 GB: 6-of-54 16.1 GB: 7-of-54 16.1 GB: 8-of-54 16.1 GB: 9-of-54 16.1 GB: 10-of-54 16.1 GB: 11-of-54 16.1 GB: 12-of-54 16.1 GB: 13-of-54 16.1 GB: 14-of-54 16.1 GB: 15-of-54 16.1 GB: 16-of-54 16.1 GB: 17-of-54 16.1 GB: 18-of-54 16.1 GB: 19-of-54 16.1 GB: 20-of-54 16.1 GB: 21-of-54 16.1 GB: 22-of-54 16.1 GB: 23-of-54 16.1 GB: 24-of-54 16.1 GB: 25-of-54 16.1 GB: 26-of-54 16.1 GB: 27-of-54 16.1 GB: 28-of-54 16.1 GB: 29-of-54 16.1 GB: 30-of-54 16.1 GB: 31-of-54 16.1 GB: 32-of-54 16.1 GB: 33-of-54 16.1 GB: 34-of-54 16.1 GB: 35-of-54 16.1 GB: 36-of-54 16.1 GB: 37-of-54 16.1 GB: 38-of-54 16.1 GB: 39-of-54 16.1 GB: 40-of-54 16.1 GB: 41-of-54 16.1 GB: 42-of-54 16.1 GB: 43-of-54 16.1 GB: 44-of-54 16.1 GB: 45-of-54
Supported Languages	ru en
Model Architecture	GigaChat35ForCausalLM
License	mit
Context Length	131072
Model Max Length	131072
Transformers Version	4.57.3
Vocabulary Size	128256

Rank the GigaChat3.5 432B A28B 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 GigaChat3.5 432B A28B Base

Best Alternatives	Context / RAM	Downloads	Likes
...igaChat3.5 432B A28B Reasoning	256K / 445.8 GB	271	31
...at3.5 432B A28B Reasoning Bf16	256K / 708.4 GB	300	6
GigaChat3.5 432B A28B	256K / 437.7 GB	884	69

Note: green Score (e.g. "73.2") means that the model is better than ai-sage/GigaChat3.5-432B-A28B-base.