We raised a $75m series C to build the future of inference

Model library

Browse our library of open source models that are ready to deploy behind an API endpoint in seconds.

Deploy your own model

LLM Transcription Text to speech Image generation Embedding

large language models

Qwen 3 4B

LLM

V3TRT-LLMH100

Qwen 3 32B

LLM

V3TRT-LLMH100

Qwen 3 235B

LLM

V3SGLangH100

Llama 4 Scout

Llama 4 Scout

LLM

V4.0InstructvLLMH100

Llama 4 Maverick

Llama 4 Maverick

LLM

V4.0InstructvLLMB200

DeepSeek-R1

LLM

R1SGLangB200

text to speech models

Orpheus TTS

Orpheus TTS

Text to speech

TRT-LLMH100 MIG 40GB

MARS6

Text to speech

V6L4

XTTS V2

Text to speech

T4

glowing gold goddess, anime style art. Kokoro from the anime Terminator

Kokoro

Text to speech

fp16T4

transcription models

Whisper (best performance)

Transcription

V3H100 MIG 40GB

WhisperX

Transcription

L4

Whisper V3

Transcription

V3H100 MIG 40GB

Whisper V3 Turbo

Transcription

V3TurboH100 MIG 40GB

image generation models

ZenCtrl

Image generation

Custom ServerH100

Prompt: A movie still of an astronaut with a dark visor with a lightning bolt superhero icon on his chest

SDXL Lightning

Image generation

1.0LightningA100

moss covered astronaut

Stable Diffusion 3 Medium

Image generation

3A100

Stable Diffusion XL

Image generation

XL 1.0A10G

flux-dev

Image generation

devbloat16H100 MIG 40GB

flux-schnell

Image generation

schnellbfloat16H100 MIG 40GB

embedding models

Tulu 3 8B Reward

Embedding

V3RewardBEIH100 MIG 40GB

BGE Reranker M3

Embedding

BEIH100

BGE Embedding ICL

Embedding

BEIH100

Mixedbread Embed Large V1

Embedding

V1EmbeddingBEIL4

Prompt: A movie still of a squirrel in a forest green ski suit

Nomic Embed Code

Embedding

BEIH100 MIG 40GB

DeepSeek models

DeepSeek-R1

LLM

R1SGLangB200

DeepSeek-V3

LLM

V3SGLangB200

DeepSeek-R1 Llama 70B

LLM

R1LlamaTRT-LLMH100

DeepSeek-R1 Qwen 32B

LLM

R1QwenTRT-LLMH100

DeepSeek-R1 Qwen 7B

LLM

R1QwenTRT-LLMH100 MIG 40GB

DeepSeek-R1 Zero

LLM

R1ZeroSGLangH200

Qwen models

Qwen 3 4B

LLM

V3TRT-LLMH100

Qwen 3 32B

LLM

V3TRT-LLMH100

Qwen 3 235B

LLM

V3SGLangH100

Qwen 2.5 14B Instruct

LLM

2.5TRT-LLMH100

Qwen 2.5 32B Coder Instruct

LLM

2.5CoderTRT-LLMH100

Qwen 2.5 7B Math Instruct

LLM

2.5MathTRT-LLMH100 MIG 40GB

Meta models

Llama 4 Scout

Llama 4 Scout

LLM

V4.0InstructvLLMH100

Llama 4 Maverick

Llama 4 Maverick

LLM

V4.0InstructvLLMB200

Llama 3.3 70B Instruct

LLM

3.3TRT-LLMH100

Llama 3.1 8B Instruct

LLM

3.1InstructTRT-LLMH100

Llama 3.1 405B Instruct

LLM

3.1InstructH100

Llama 11B Vision

Llama 3.2 11B Vision Instruct

LLM

3.2VisionA100

Deploy any model in just a few commands

Avoid getting tangled in complex deployment processes. Deploy best-in-class open-source models and take advantage of optimized serving for your own models.

Start deploying

$

truss init -- example stable-diffusion-2-1-base ./my-sd-truss

$

cd ./my-sd-truss

$

export BASETEN_API_KEY=MdNmOCXc.YBtEZD0WFOYKso2A6NEQkRqTe

$

truss push

INFO

Serializing Stable Diffusion 2.1 truss.

INFO

Making contact with Baseten 👋 👽

INFO

🚀 Uploading model to Baseten 🚀

Upload progress: 0% | | 0.00G/2.39G