Predibase

Predibase is a platform for fine-tuning and serving open-source LLMs. It pairs efficient LoRA / Turbo LoRA supervised and reinforcement (GRPO) fine-tuning with serverless and dedicated inference powered by LoRAX, the open-source multi-LoRA serving stack that packs hundreds of adapters onto a single GPU. Inference is exposed through an OpenAI-compatible API plus native generate endpoints.

8 APIs 0 Features

AILLMFine-TuningInferenceLoRA

APIs

Predibase Inference (OpenAI-Compatible) API

OpenAI-compatible chat completions and completions served from Predibase serverless and dedicated deployments, with per-request LoRA adapter selection via the model field and SS...

Predibase Prompt / Generate API

Native text-generation endpoints (generate and generate_stream) for prompting deployed base models and fine-tuned adapters, with adapter source selection (pbase, hub, or s3) and...

Predibase Fine-Tuning API

Create and manage supervised and reinforcement (GRPO) fine-tuning jobs that train efficient LoRA / Turbo LoRA adapters on top of open-source base models, returning adapter versi...

Predibase Adapters API

Manage adapter repositories and the trained adapter versions inside them - the LoRA artifacts produced by fine-tuning jobs that are loaded onto deployments for inference.

Predibase Deployments API

Create, read, update, and delete dedicated and private serverless deployments, selecting a base model and GPU accelerator (A10, A100) and enabling LoRA serving for fine-tuned ad...

Predibase Datasets API

Connect and manage datasets used as input to fine-tuning jobs, uploaded from files or referenced from connected storage.

Predibase Models API

List the open-source base models supported on Predibase for fine-tuning and serving, with metadata used when creating jobs and deployments.

Predibase Batch Inference API

Launch asynchronous batch inference jobs against a base model with per-row adapter selection, billed at a flat per-million-token batch rate for non-realtime workloads.

Event Specifications

Predibase Inference Streaming (HTTP + SSE)

AsyncAPI 2.6 description of Predibase's **inference streaming** surface. Predibase does not publish a WebSocket API. The only asynchronous / event-style transport documented at ...

ASYNCAPI

Resources

👥

GitHubOrganization

Sources

aid: predibase
url: https://raw.githubusercontent.com/api-evangelist/predibase/refs/heads/main/apis.yml
name: Predibase
kind: company
description: Predibase is a platform for fine-tuning and serving open-source LLMs.
  It pairs efficient LoRA / Turbo LoRA supervised and reinforcement (GRPO) fine-tuning
  with serverless and dedicated inference powered by LoRAX, the open-source multi-LoRA
  serving stack that packs hundreds of adapters onto a single GPU. Inference is
  exposed through an OpenAI-compatible API plus native generate endpoints.
image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg
tags:
- AI
- LLM
- Fine-Tuning
- Inference
- LoRA
created: '2026-06-20'
modified: '2026-06-20'
specificationVersion: '0.19'
apis:
- aid: predibase:predibase-inference-openai-api
  name: Predibase Inference (OpenAI-Compatible) API
  tags:
  - Chat
  - Completions
  - LLM
  - OpenAI Compatible
  image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg
  humanURL: https://docs.predibase.com/user-guide/inference/migrate-openai
  baseURL: https://serving.app.predibase.com/{tenant}/deployments/v2/llms/{model}/v1
  properties:
  - url: https://docs.predibase.com/user-guide/inference/migrate-openai
    type: Documentation
  - url: https://docs.predibase.com/user-guide/inference/rest_api
    type: APIReference
  - url: openapi/predibase-openapi.yml
    type: OpenAPI
  - url: asyncapi/predibase-asyncapi.yml
    type: AsyncAPI
  - url: collections/predibase.postman_collection.json
    type: PostmanCollection
  description: OpenAI-compatible chat completions and completions served from
    Predibase serverless and dedicated deployments, with per-request LoRA adapter
    selection via the model field and SSE streaming when stream is true.
- aid: predibase:predibase-prompt-generate-api
  name: Predibase Prompt / Generate API
  tags:
  - Prompt
  - Generate
  - Streaming
  image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg
  humanURL: https://docs.predibase.com/user-guide/inference/rest_api
  baseURL: https://serving.app.predibase.com/{tenant}/deployments/v2/llms/{model}
  properties:
  - url: https://docs.predibase.com/user-guide/inference/rest_api
    type: Documentation
  - url: https://docs.predibase.com/user-guide/inference/querying-models/text-generation
    type: APIReference
  - url: openapi/predibase-openapi.yml
    type: OpenAPI
  - url: collections/predibase.postman_collection.json
    type: PostmanCollection
  description: Native text-generation endpoints (generate and generate_stream) for
    prompting deployed base models and fine-tuned adapters, with adapter source
    selection (pbase, hub, or s3) and token-streaming responses.
- aid: predibase:predibase-fine-tuning-api
  name: Predibase Fine-Tuning API
  tags:
  - Fine-Tuning
  - LoRA
  - GRPO
  - Reinforcement Learning
  image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg
  humanURL: https://docs.predibase.com/user-guide/fine-tuning/overview
  baseURL: https://api.app.predibase.com/v2
  properties:
  - url: https://docs.predibase.com/user-guide/fine-tuning/overview
    type: Documentation
  - url: https://docs.predibase.com/user-guide/fine-tuning/grpo
    type: Documentation
  - url: openapi/predibase-openapi.yml
    type: OpenAPI
  - url: collections/predibase.postman_collection.json
    type: PostmanCollection
  description: Create and manage supervised and reinforcement (GRPO) fine-tuning
    jobs that train efficient LoRA / Turbo LoRA adapters on top of open-source base
    models, returning adapter versions for serving.
- aid: predibase:predibase-adapters-api
  name: Predibase Adapters API
  tags:
  - Adapters
  - Repos
  - Versions
  image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg
  humanURL: https://docs.predibase.com/fine-tuning/adapters
  baseURL: https://api.app.predibase.com/v2
  properties:
  - url: https://docs.predibase.com/fine-tuning/adapters
    type: Documentation
  - url: openapi/predibase-openapi.yml
    type: OpenAPI
  - url: collections/predibase.postman_collection.json
    type: PostmanCollection
  description: Manage adapter repositories and the trained adapter versions inside
    them - the LoRA artifacts produced by fine-tuning jobs that are loaded onto
    deployments for inference.
- aid: predibase:predibase-deployments-api
  name: Predibase Deployments API
  tags:
  - Deployments
  - Dedicated
  - Serverless
  image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg
  humanURL: https://docs.predibase.com/user-guide/inference/dedicated_deployments
  baseURL: https://api.app.predibase.com/v2
  properties:
  - url: https://docs.predibase.com/user-guide/inference/dedicated_deployments
    type: Documentation
  - url: https://docs.predibase.com/user-guide/inference/private_deployments
    type: Documentation
  - url: openapi/predibase-openapi.yml
    type: OpenAPI
  - url: collections/predibase.postman_collection.json
    type: PostmanCollection
  description: Create, read, update, and delete dedicated and private serverless
    deployments, selecting a base model and GPU accelerator (A10, A100) and
    enabling LoRA serving for fine-tuned adapters.
- aid: predibase:predibase-datasets-api
  name: Predibase Datasets API
  tags:
  - Datasets
  - Training Data
  image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg
  humanURL: https://docs.predibase.com/user-guide/getting-started/fine-tuning-and-serving
  baseURL: https://api.app.predibase.com/v2
  properties:
  - url: https://docs.predibase.com/user-guide/getting-started/fine-tuning-and-serving
    type: Documentation
  - url: openapi/predibase-openapi.yml
    type: OpenAPI
  - url: collections/predibase.postman_collection.json
    type: PostmanCollection
  description: Connect and manage datasets used as input to fine-tuning jobs,
    uploaded from files or referenced from connected storage.
- aid: predibase:predibase-models-api
  name: Predibase Models API
  tags:
  - Models
  - Catalog
  - Base Models
  image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg
  humanURL: https://docs.predibase.com/user-guide/inference/models
  baseURL: https://api.app.predibase.com/v2
  properties:
  - url: https://docs.predibase.com/user-guide/inference/models
    type: Documentation
  - url: openapi/predibase-openapi.yml
    type: OpenAPI
  - url: collections/predibase.postman_collection.json
    type: PostmanCollection
  description: List the open-source base models supported on Predibase for
    fine-tuning and serving, with metadata used when creating jobs and deployments.
- aid: predibase:predibase-batch-inference-api
  name: Predibase Batch Inference API
  tags:
  - Batch
  - Async
  image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg
  humanURL: https://docs.predibase.com/user-guide/inference/BatchInference
  baseURL: https://api.app.predibase.com/v2
  properties:
  - url: https://docs.predibase.com/user-guide/inference/BatchInference
    type: Documentation
  - url: openapi/predibase-openapi.yml
    type: OpenAPI
  - url: collections/predibase.postman_collection.json
    type: PostmanCollection
  description: Launch asynchronous batch inference jobs against a base model with
    per-row adapter selection, billed at a flat per-million-token batch rate for
    non-realtime workloads.
common:
- type: GitHubOrganization
  url: https://github.com/predibase
- type: LinkedIn
  url: https://www.linkedin.com/company/predibase
- type: Website
  url: https://predibase.com
- type: Documentation
  url: https://docs.predibase.com
- type: Plans
  url: plans/predibase-plans-pricing.yml
- type: RateLimits
  url: rate-limits/predibase-rate-limits.yml
- type: FinOps
  url: finops/predibase-finops.yml
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com