Models / OpenAI

OpenAI

Deploy gpt-oss-120B and gpt-oss-20B on Together AI. Frontier reasoning performance under Apache 2.0 with complete model ownership.

Why OpenAI on Together AI?

Designed for production workloads that need 
consistent performance and operational control.

Frontier reasoning, open license

gpt-oss-120B and gpt-oss-20B deliver o3-class reasoning performance under Apache 2.0, with no restrictions on commercial use, fine-tuning, or deployment.

Deploy anywhere, own everything

Air-gapped deployments, on-premises, or Together AI cloud. Full model ownership means your infrastructure, your data, your terms.

Enterprise infrastructure from day one

99.9% uptime SLA, multi-region deployment, SOC 2 Type II certified, and HIPAA compliant. North American infrastructure with US-based deployment.

Meet the OpenAI family

Explore top-performing models across text, image, video, code, and voice.

New

Chat

gpt-oss-120B

new

Image

GPT Image 1.5

New

Video

Sora 2 Pro

Transcribe

Whisper Large v3

Video

Sora 2

Transcribe

Whisper Large v3 (Streaming)

New

Code

gpt-oss-20B

Breakthrough technical innovations

Explore all the game-changing architectural advances that make OpenAI models shine.

  • Mixture of Experts (MoE)

    Sparse expert routing activates only 37B out of 671B parameters for each token in V3. Advanced load balancing without auxiliary losses maintains performance while reducing computational cost.

  • Group Relative Policy Optimization

    New RL approach that removes separate value networks in RLHF, using grouped relative advantage estimation to cut compute requirements while maintaining training stability.

  • Native Reasoning Transparency

    First reasoning model to expose complete thinking process in <think> tags. Native reasoning capabilities built into model foundation through large-scale reinforcement learning.

  • MetaP Training

    First successful implementation of FP8 mixed precision training on a 671B parameter model. Pioneering reinforcement learning approach without supervised fine-tuning as preliminary step.

  • Multi-Head Latent Attention

    Innovative attention mechanism that reduces KV-cache memory requirements while maintaining modeling performance. Optimized for efficient inference deployment.

  • Multi-Token Prediction

    Novel training objective that allows the model to predict multiple tokens simultaneously. Enhanced performance and efficiency through advanced training techniques.

Deployment options

Run models using different deployment options depending on latency needs, traffic patterns, and infrastructure control.

  • Serverless

  • Inference

Serverless Inference

Real-time

A fully managed inference API that automatically scales with request volume.

Best for

Variable or unpredictable traffic

Rapid prototyping and iteration

Cost-sensitive or early-stage production workloads

Batch

Process massive workloads of up to 30 billion tokens asynchronously, at up to 50% less cost.

Best for

Classifying large datasets

Offline summarization

Synthetic data generation

Dedicated Inference

Dedicated Model Inference

An inference endpoint backed by reserved, isolated compute resources and the Together AI inference engine.

Best for

Predictable or steady traffic

Latency-sensitive applications

High-throughput production workloads

Dedicated Container Inference

Run inference with your own engine and model on fully-managed, scalable infrastructure.

Best for

Generative media models

Non-standard runtimes

Custom inference pipelines