From the model card
Fields below are copied from the tags and counters on the HuggingFace repository TinyLlama/TinyLlama-1.1B-Chat-v1.0 at our last fetch. They are set by the uploader, not verified by us; rows with no tag are omitted. How this page is made.
- Publisher (HF namespace)
- TinyLlama
- Pipeline tag
- text-generation
- Library
- Transformers
- Weight formats
- safetensors
- License tag
apache-2.0— read the license file in the repo before relying on it- Language tags
- English (en)
- Datasets declared
- cerebras/SlimPajama-627B, bigcode/starcoderdata, HuggingFaceH4/ultrachat_200k, HuggingFaceH4/ultrafeedback_binarized
- Downloads (HF counter at last fetch)
- 1,749,916
- Likes (HF counter at last fetch)
- 1,764
- Model card
- https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v1.0
Use cases
- Local chat assistant on CPU-only or low-RAM hardware
- Rapid prototyping of LLM application logic
- Edge deployment scenarios where 7B+ models are infeasible
- Lightweight intent parsing in structured output pipelines
Pros
- ~2GB memory footprint in float16
- Trained for 3 trillion tokens — more data than many larger models
- Apache-2.0 licensed
- Wide GGUF quantization support via llama.cpp
Cons
- 1.1B parameters produce noticeably weaker reasoning than 7B models
- Hallucination rate is high on factual queries
- Knowledge cutoff from mid-2023 training data
- Instruction following quality degrades on multi-step or complex constraints
Tags
transformerssafetensorsllamatext-generationconversationalendataset:cerebras/SlimPajama-627Bdataset:bigcode/starcoderdatadataset:HuggingFaceH4/ultrachat_200kdataset:HuggingFaceH4/ultrafeedback_binarizedlicense:apache-2.0text-generation-inferenceendpoints_compatibleregion:usdeploy:sagemakerdeploy:azure