This is a very small test model derived from the DeepSeek V3 architecture, published under the trl-internal-testing organization. It is intended for internal validation of training, inference, and fine-tuning code within the Hugging Face TRL ecosystem rather than for production use. The model includes tokenizer configurations compatible with common chat templates.
Tiny DeepseekV3ForCausalLM is a Foundation models & chat product. It focuses on providing a minimal model for testing and validating the TRL (Transformers Reinforcement Learning) library and training pipelines. It is built as an open-source project for developers. Tiny DeepseekV3ForCausalLM is open source under the Apache-2.0 license. Tiny DeepseekV3ForCausalLM is available on the web, the command line, and API.
Behind Tiny DeepseekV3ForCausalLM is trl-internal-testing, and the product first shipped in 2020. Development happens publicly on GitHub with 18.9k stars and 523 commits in the last 90 days.
Latest indexed changes and source events
trl-internal-testing/tiny-DeepseekV3ForCausalLM verified by the PulseGate indexer
