ai-sage / text-generation updated 2 months ago

GigaChat3.1-Audio-10B-A1.8B

Name: GigaChat3.1-Audio-10B-A1.8B
Author: ai-sage

GigaChat Audio 10B is an audio-native LLM built on top of the GigaChat 3.1 Lightning text model. A Conformer speech encoder and a modality adapter feed audio embeddings directly into a Mixture-of-Experts decoder, so the model keeps the text quality of its base while adding speech understanding.

Params

—

Context

—

Downloads 30d

331 K

Likes

Commercial use: allowed mit Not gated SAFETENSORS 2 languages View on Hugging Face ↗

Download history

daily snapshots · 44 days

▲ 147 K in the last 30 days (80.0%)

335 K165 K

Aug 8Aug 22Sep 6Sep 20

Can you run it?

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

File	Quant	Size	Est. VRAM	Verdict on RTX 4090 · 24 GB
model.safetensors	fp16	23.7 GB	26.6 GB	❌ Won’t fit

Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

Run it

copy-paste, exact tags checked against the Hub

~ · curl · api/v1

$ curl -s https://aimodelscomparison.com/api/v1/models/gigachat3-1-audio-10b-a1-8b
{
  "hf_id": "ai-sage/GigaChat3.1-Audio-10B-A1.8B",
  "params_b": null,
  "context_length": null,
  "license": { "id": "mit", "commercial": "yes" },
  "downloads_30d": 331017,
  "vram_estimates": [
    { "quant": "fp16", "gb": 26.6 }
  ],
  "updated_at": "2026-08-08T01:00:30Z"
}

est. VRAM —on RTX 4090 · 24 GBJSON API →

Specifications

Architecture: GigaChatAudioForConditionalGeneration
Licence: mit
First seen on the Hub: 2026-07-13
Training datasets: undisclosed
Added to our catalog: 2026-08-08

Compare with any text-generation model