Managed service for fine-tuning open-source models on dedicated or shared GPU infrastructure for production use.
Together AI Fine-Tuning is a managed service that lets teams fine-tune open-source large language models on dedicated or shared GPU infrastructure.
It supports supervised fine-tuning to improve accuracy, reduce hallucinations, and steer model behavior, with trained models deployable for production inference.
Chonkie
Model Specialization Service
chonkie.ai
Thinking Machines Lab
Tinker
thinkingmachines.ai
Baseten
Training on Baseten
baseten.co
| Attribute | Together AI Fine-Tuning | Model Specialization Service | Tinker | Training on Baseten | |
|---|---|---|---|---|---|
| Provider |  | | | | |
| Founded | 2022 | 2025 | 2025 | 2019 | |
| Sells To | B2B, Enterprise | Enterprises | — | Enterprise | |
| Pricing Model | Recurring, Subscription, Usage-based | Service Fees | — | Recurring, Software, Usage-based | |
| Ownership | Venture-backed | Venture-backed | Venture-backed | Venture-backed |
Provider
Founded 2022
Sells To B2B, Enterprise
Pricing Model Recurring, Subscription, Usage-based
Ownership Venture-backed
Provider
Founded 2025
Sells To Enterprises
Pricing Model Service Fees
Ownership Venture-backed
Provider
Founded 2025
Sells To —
Pricing Model —
Ownership Venture-backed
Provider
Founded 2019
Sells To Enterprise
Pricing Model Recurring, Software, Usage-based
Ownership Venture-backed
Managed supervised fine-tuning service for open-source models with on-demand and dedicated deployment options.
Async inference API for long-horizon AI agents, serving leading open-source models through OpenAI- and Anthropic-compatible endpoints.
A unified OpenAI-compatible API and gateway for accessing hundreds of LLMs from many providers.
RunInfra Model APIs provide hosted open-model inference through compatible APIs for developers and agent teams.
Replicate is a cloud API to run, fine-tune, and deploy open-source machine learning models with auto-scaling.
Secure enterprise portal for multiple large language model platforms with reusable language skills and usage-based pricing.
Developer API for Anthropic's Claude family of frontier language models including Opus, Sonnet, and Haiku.
An integrated compute, training, inference, and sandbox stack for enterprises to train and deploy their own AI models.
Open source AI gateway routing every model behind one zero-markup OpenAI-compatible endpoint.
An open-source AI gateway and proxy that calls many large language model APIs in the OpenAI format.
Developer API providing pay-per-token access to DeepSeek's hosted language models.
Managed supervised fine-tuning service for open-source models with on-demand and dedicated deployment options.
Async inference API for long-horizon AI agents, serving leading open-source models through OpenAI- and Anthropic-compatible endpoints.
A unified OpenAI-compatible API and gateway for accessing hundreds of LLMs from many providers.
Adaptive OpenAI-compatible endpoint coordinating open-weight models for chat and agents.
RunInfra Model APIs provide hosted open-model inference through compatible APIs for developers and agent teams.
Replicate is a cloud API to run, fine-tune, and deploy open-source machine learning models with auto-scaling.
Secure enterprise portal for multiple large language model platforms with reusable language skills and usage-based pricing.
Developer API for Anthropic's Claude family of frontier language models including Opus, Sonnet, and Haiku.
An integrated compute, training, inference, and sandbox stack for enterprises to train and deploy their own AI models.
Open source AI gateway routing every model behind one zero-markup OpenAI-compatible endpoint.
An open-source AI gateway and proxy that calls many large language model APIs in the OpenAI format.
Developer API providing pay-per-token access to DeepSeek's hosted language models.