TinyLlama Models by TinyLlama
Last refreshed 2026-05-19. Next refresh: weekly.
Details
Capabilities
About
TinyLlama is an open project from the StatNLP Research Group at Singapore University of Technology and Design that pretrained a compact 1.1B-parameter Llama model on 3 trillion tokens. It uses exactly the same architecture and tokenizer as Llama 2, so it drops into many open-source projects built for Llama, and its small size suits apps with tight compute and memory budgets. The team published intermediate checkpoints during training. TinyLlama-1.1B-Chat-v1.0 is the chat model, fine-tuned from the 3-trillion-token checkpoint with Hugging Face's Zephyr recipe: first on a variant of the UltraChat dialogue dataset, then aligned with DPO on UltraFeedback.
Decision facts
- Best fit
- structured outputs
- Capability starting point
- Together AI TinyLlama-1.1B-Chat-v1.0 with 2K context and structured outputs
- Lowest tracked input
- Together AI TinyLlama-1.1B-Chat-v1.0 · $0.050/1M · Together AI
- Closest related family
- 360Zhinao
Current Variants
Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.
Use when the workload needs 2K context, 1.1B parameters, and structured outputs.
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Together AI TinyLlama-1.1B-Chat-v1.0 | Use when the workload needs 2K context, 1.1B parameters, and structured outputs. | 2024-01 | 2K context1.1B parametersstructured outputs | Current |
Release Timeline
1 release groupSpecifications(1 models)
| Model | Released | Context | Parameters | Structured Outputs |
|---|---|---|---|---|
| Together AI TinyLlama-1.1B-Chat-v1.0 | 2024-01 | 2K | 1.1B | Yes |
Available From(1 provider)
Pricing
| Model | Provider | Input / 1M | Output / 1M | Type |
|---|---|---|---|---|
| Together AI TinyLlama-1.1B-Chat-v1.0 | Together AI | $0.05 | $0.05 | Serverless |