Model ID
CohereLabs/tiny-aya-base
Capabilities
Tiny Aya Base
Tiny Aya Base is a 3.35B-parameter multilingual language model designed for efficient deployment and broad language coverage. It uses a 36-layer transformer with 2,048 hidden size, 16 attention heads, and grouped KV heads (4 KV heads). The architecture alternates sliding window attention (4,096 window) with periodic global attention layers for full-sequence interaction. It supports around 8K context in practice and uses RoPE for positional encoding. Trained across 70+ languages, it is optimized for balanced multilingual performance, downstream adaptation, and low-resource environments.
About this model
Tiny Aya Base is a 3.35B-parameter multilingual language model designed for efficient deployment and broad language coverage. It uses a 36-layer transformer with 2,048 hidden size, 16 attention heads, and grouped KV heads (4 KV heads). The architecture alternates sliding window attention (4,096 window) with periodic global attention layers for full-sequence interaction. It supports around 8K context in practice and uses RoPE for positional encoding. Trained across 70+ languages, it is optimized for balanced multilingual performance, downstream adaptation, and low-resource environments.
Model specifications and capabilities are published by the model author and reproduced here from HF Model Card (CohereLabs/tiny-aya-base). Released under CC-BY-NC-4.0. Further reading: Paper · Blog
Deploy Private Inference Endpoint
Choose your hardware and inference engine to get deployment commands and performance benchmarks tailored to your infrastructure
Hardware
MI325X
2,048 GB
B200
1,440 GB
Engine
vLLM
SGLang
Deployment code
Access Restricted
Enter your email to get access to this content
Download the chat template before serving: wget "curl -L https://raw.githubusercontent.com/vllm-project/vllm/main/examples/template_chatml.jinja \ -o template_chatml.jinja" and pass --chat-template "/tmp/template_chatml.jinja"
Performance under load
Tokens/s (total)
Benchmarks measured by Vultr on MI325X · vLLM. Throughput and latency vary with concurrency, input length and engine configuration, so treat these as a comparison baseline, not a service guarantee.
AMD Instinct MI325X
Unlock Next-Generation AI Power on Vultr Cloud GPU