GigaChat3.1 Audio 10B A1.8B — Hardware Requirements & GPU Compatibility
ChatFunctions
GigaChat3.1 Audio 10B A1.8B is a 10B-parameter open language model from ai-sage. It supports a context window of up to 262,144 tokens. At BF16 it needs about 20.63 GB of VRAM — see which GPUs and Macs can run it below.
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run GigaChat3.1 Audio 10B A1.8B
One recommended way. Every other method, OS and quant is in the step-by-step guide.
Recommended
vllm serve ai-sage/GigaChat3.1-Audio-10B-A1.8B --max-model-len 16384vLLM repo with its own quantisation
Download: Weights on Hugging Face
Other ways
each opens step-by-stepRun in cloud
Fits on RTX A6000 (48 GB) (25 GB headroom) · BF16
- Generation speed
- ~162 tok/s
- generation speed
- Cost per 1M output tokens
- $0.57
- per 1M output tokens
How Much VRAM Does GigaChat3.1 Audio 10B A1.8B Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 20.6 GB | 25.5 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Fits: 16K context · ~198 tok/s |
Selected BF16
Brain floating point 1622.9GBat 16K context
- Weights
- 20.0 GB
- KV cache
- 2.6 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 1.1 GB free · ~198 tok/s
≥50 tok/s needs ≥197 GB/s · RTX 5090 ~288, RTX 3090 Ti ~198
Which Hardware Can Run GigaChat3.1 Audio 10B A1.8B?
BF16 · 20.6 GBGigaChat3.1 Audio 10B A1.8B (BF16) requires 20.6 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 27+ GB is recommended. Using the full 262K context window can add up to 41.6 GB, bringing total usage to 62.2 GB. 7 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 100 devices with unified memory can run GigaChat3.1 Audio 10B A1.8B, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), Mac Studio M1 Max (32 GB).
≤ 20 GB
≤ 20 GB tier →doesn't fit · smallest quant needs 22.9 GB
32–40 GB
fits · 9.1 GB free · up to 64K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~288Simulate output on RTX 5090M5 Max36 GB~80Simulate output on Mac Studio M5 Max (36 GB)M532 GB~37Simulate output on MacBook Air 13" M5 (32 GB)Orin~52Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~34Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at BF16
Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Qwen3.5 9BAlibaba | 99.4 | 6.6 GB | ~103 | same size |
| Ornith 1.5 9Bornith-ai | — | 6.6 GB | ~103 | same size |
| MiniCPM O 4 5OpenBMB | — | 8.3 GB | ~105 | same size |
| GLM 4.1V 9B ThinkingZ.ai | — | 7.2 GB | ~100 | same size |
| Qwen2.5 Omni 7BAlibaba | — | 9.9 GB | ~93 | same size |
| Llama 3.2 11B Vision InstructMeta | — | 9.8 GB | ~93 | same size |
Specifications
- Publisher
- ai-sage
- Parameters
- 10B
- Active parameters
- 1.2B per token
- Architecture
- GigaChatAudioForConditionalGeneration
- Context Length
- 262,144 tokens
- Vocabulary Size
- 128,256
- Release Date
- 2026-07-13
- License
- MIT
Frequently Asked Questions
How much VRAM does GigaChat3.1 Audio 10B A1.8B need?
GigaChat3.1 Audio 10B A1.8B requires 20.6 GB of VRAM at BF16. Full 262K context adds up to 41.5 GB (62.2 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 10B × 16 bits ÷ 8 = 20 GB
KV Cache + Overhead ≈ 0.6 GB (at 2K context + ~0.3 GB framework)
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
KV Cache + Overhead ≈ 42.2 GB (at full 262K context)
VRAM usage by quantization
Can I run GigaChat3.1 Audio 10B A1.8B on a Mac?
Yes — Mac Studio M1 Max (32 GB) and 84 other Macs can run GigaChat3.1 Audio 10B A1.8B. Apple Silicon uses unified memory, so the model shares RAM with the system. At BF16 you need at least 20.6 GB of usable unified memory (RAM minus macOS overhead).
Can I run GigaChat3.1 Audio 10B A1.8B locally?
Yes — GigaChat3.1 Audio 10B A1.8B can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is GigaChat3.1 Audio 10B A1.8B?
At BF16, GigaChat3.1 Audio 10B A1.8B can reach ~176 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~198 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = 1000 ÷ (active GB ÷ (bandwidth GB/s × efficiency) × 1000 + layers × routing ms)
Mixture-of-Experts: just the active experts get read for each token, along with a fixed routing cost per layer.
Example: NVIDIA B200 → 2.4 GB active ÷ (8000 × 0.65) = 0.457 ms, plus 26 layers × 0.055 ms = 1.430 ms, so 1000 ÷ 1.887 ms = ~530 tok/s
Estimated speed at BF16 (20.6 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of GigaChat3.1 Audio 10B A1.8B?
At BF16, the download is about 20.00 GB.
Which GPUs can run GigaChat3.1 Audio 10B A1.8B?
7 consumer GPUs can run GigaChat3.1 Audio 10B A1.8B at BF16. Top options include NVIDIA GeForce RTX 5090, AMD Radeon RX 7900 XTX, NVIDIA GeForce RTX 3090. 1 GPU has plenty of headroom for comfortable inference.
Which devices can run GigaChat3.1 Audio 10B A1.8B?
100 devices with unified memory can run GigaChat3.1 Audio 10B A1.8B at BF16 (20.6 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.