Qwen3.5 0.8B (Q8_0) — 8 GBon Scaleway H100-1-80G
Overview
Qwen3.5 0.8B is a 0.87B parameter dense language model by Qwen, with code, multilingual, thinking, tool-calls, vision capabilities. It supports a context window of up to 262,144 tokens.
Qwen3.5 0.8B is the smallest model in Alibaba's Qwen 3.5 family, built on the Gated Delta Networks hybrid architecture with 0.87 billion parameters, purpose-built for phones, edge devices, and ultra-constrained environments. It is natively multimodal, processing text, images, and video, with built-in thinking capabilities for chain-of-thought reasoning. The model supports a 262K context window and covers over 201 languages. Released under the Apache 2.0 license, it quantizes down to under 1 GB of VRAM at Q4, making it ideal for classification and simple tasks in self-hosted deployment scenarios.
At Q8_0 quantization (high quality tier), the model weighs 0.76 GB. This fits within the 80 GB of VRAM on Scaleway H100-1-80G, enabling full GPU inference.
The NVIDIA H100 80 GB is a datacenter GPU with 80 GB of HBM3 VRAM and 3350 GB/s memory bandwidth. It delivers 1979 FP16 TFLOPS on the Hopper architecture, making it the fastest single-GPU option for large language model inference. It handles models up to 70B parameters with high throughput. Built for datacenter teams running demanding production AI workloads.
Hardware Requirements
| Model size | 0.76 GB |
| VRAM available | 80 GB |
| VRAM used | 8 GB |
| System RAM | 240 GB |
| Min RAM required | 0 GB |
| GPU layers | 24 / 24 |
| Context size | 262,144 |
| Backend | cuda13 |
| Flash attention | Yes |
| Reading from disk | Yes |
Deploy
Prerequisites
Ensure your GPU nodes are prepared with the NVIDIA container toolkit:
ansible-playbook prositronic.infra.nvidia_container_toolkitCommand
helmfile --state-values-file <(curl -s https://www.prositronic.eu/values/qwen3-5-0-8b/q8_0/nvidia-h100-80gb.yaml) applyGenerated values.yaml
/values/qwen3-5-0-8b/q8_0/nvidia-h100-80gb.yaml
Loading values…Frequently Asked Questions
How much VRAM does Qwen3.5 0.8B (Q8_0) need?
The Q8_0 quantization of Qwen3.5 0.8B requires 0.76 GB. All 24 layers fit in the 80 GB of VRAM available on Scaleway H100-1-80G, enabling full GPU acceleration.
Can I run Qwen3.5 0.8B on Scaleway H100-1-80G?
Yes. Scaleway H100-1-80G provides 80 GB of VRAM, which is enough to run Qwen3.5 0.8B (Q8_0) with all layers on the GPU for optimal performance.
What is quantization?
Quantization reduces a model's numerical precision from its original floating-point format to a more compact representation. This shrinks the file size and VRAM footprint, making it possible to run large models on consumer hardware. The trade-off is a small reduction in output quality. Q8_0 compresses Qwen3.5 0.8B from its original size down to 0.76 GB.
What quantization should I choose for Qwen3.5 0.8B?
Q8_0 is a high-quality quantization. Higher-quality quants (Q8, Q6) preserve more model accuracy but need more VRAM. Lower quants (Q4, Q3, Q2) reduce VRAM usage at the cost of some quality. Choose based on your available hardware and quality requirements.
What is flash attention and why is it enabled?
Flash attention is a memory-efficient algorithm that speeds up the attention mechanism in transformer models. It reduces VRAM usage during inference by avoiding the materialisation of the full attention matrix. For Qwen3.5 0.8B on Scaleway H100-1-80G, flash attention is enabled to maximise context length and throughput within the available 80 GB of VRAM.
How do I run Qwen3.5 0.8B (Q8_0) with Ollama?
Run ollama run qwen3.5:0.8b-q8_0 to start Qwen3.5 0.8B (Q8_0). Ollama handles downloading the model weights automatically on first run.