Text Generation

Decoder, encoder-decoder, recurrent, and diffusion-style text generation.

Classification follows the Hugging Face task taxonomy. See the Hugging Face Text Generation task page for the ecosystem-level task definition.

Model families

Model family	Declared recipes	Exact checkpoint examples	Runtime CLI
`bart`	2	`facebook/bart-base`	`trtmc run`
`bloom`	2	`bigscience/bloom-560m`	`trtmc run`
`codegen`	2	`Salesforce/codegen-350M-mono`	`trtmc run`
`deepseek_v2`	4	`deepseek-ai/DeepSeek-V2-Lite` `katuni4ka/tiny-random-deepseek-v3`	`trtmc run`
`elf_flow`	3	`embedded-language-flows/ELF-B-de-en` `embedded-language-flows/ELF-B-owt` `embedded-language-flows/ELF-B-xsum`	`trtmc run`
`falcon`	2	`tiiuae/falcon-rw-1b`	`trtmc run`
`gemma`	2	`google/gemma-2-2b-it`	`trtmc run`
`glm`	3	`THUDM/glm-4-9b-hf`	`trtmc run`
`gpt_neo`	2	`EleutherAI/gpt-neo-125m`	`trtmc run`
`gpt_neox`	2	`EleutherAI/pythia-70m`	`trtmc run`
`gpt_oss`	3	`openai/gpt-oss-20b`	`trtmc run`
`gpt2`	3	`distilbert/distilgpt2` `openai-community/gpt2`	`trtmc run`
`granite`	2	`ibm-granite/granite-3.1-2b-base`	`trtmc run`
`internlm`	2	`internlm/internlm2-math-plus-1_8b`	`trtmc run`
`llama`	7	`TinyLlama/TinyLlama-1.1B-Chat-v1.0` `nvidia/Llama-3.1-Minitron-4B-Depth-Base` `nvidia/Llama-3.1-Minitron-4B-Width-Base` `nvidia/Llama-3.1-Nemotron-Nano-4B-v1.1` `tiiuae/Falcon3-1B-Base`	`trtmc run`
`mamba`	2	`state-spaces/mamba-130m-hf`	`trtmc run`
`mistral`	2	`mistralai/Mistral-7B-Instruct-v0.1`	`trtmc run`
`mixtral`	2	`ggml-org/stories15M_MOE`	`trtmc run`
`nemotron`	2	`nvidia/Nemotron-4-Mini-Hindi-4B-Base` `nvidia/Nemotron-Mini-4B-Instruct`	`trtmc run`
`nemotron_h`	2	`nvidia/NVIDIA-Nemotron-Nano-9B-v2`	`trtmc run`
`nemotron_labs_diffusion`	2	`nvidia/Nemotron-Labs-Diffusion-8B`	`trtmc run`
`olmo`	2	`allenai/OLMo-1B-hf`	`trtmc run`
`olmo2`	2	`allenai/OLMo-2-0425-1B`	`trtmc run`
`opt`	2	`facebook/opt-125m`	`trtmc run`
`phi`	2	`microsoft/Phi-3-mini-4k-instruct`	`trtmc run`
`phi_moe`	3	`microsoft/Phi-tiny-MoE-instruct`	`trtmc run`
`qwen`	10	`Qwen/Qwen3-0.6B` `Qwen/Qwen3-4B-Instruct-2507`	`trtmc run`
`qwen_moe`	3	`Qwen/Qwen3-30B-A3B` `amd-quark/tiny-random-qwen3_moe`	`trtmc run`
`qwen3_5`	1	`Qwen/Qwen3.5-9B`	`trtmc run`
`rwkv`	2	`RWKV/rwkv-4-169m-pile`	`trtmc run`
`stablelm`	2	`stabilityai/stablelm-2-1_6b`	`trtmc run`
`starcoder2`	2	`bigcode/starcoder2-3b`	`trtmc run`
`t5`	2	`google-t5/t5-small`	`trtmc run`
`xglm`	2	`facebook/xglm-564M`	`trtmc run`