GPT-5.3 Chat
GPT-5.3 Chat is a conversational variant in the GPT-5.3 generation, optimized for tone, relevance, and conversational flow with reduced hallucination rates and fewer unnecessary refusals.
import { streamText } from 'ai'
const result = streamText({ model: 'openai/gpt-5.3-chat', prompt: 'Why is the sky blue?'})Playground
Try out GPT-5.3 Chat by OpenAI. Usage is billed to your team at API rates. Free users (those who haven't made a payment) get $5 of credits every 30 days.
Providers
Route requests across multiple providers. Copy a provider slug to set your preference. Visit the docs for more info. Using a provider means you agree to their terms, listed under Legal.
| Provider |
|---|
 OpenAI | 128K | 1.1s | 88tps | $1.75/M | $14.00/M | Read:$0.17/M Write:— | $10.00/K + input costs | — | +4 | 03/03/2026 |
P50 throughput on live AI Gateway traffic, in tokens per second (TPS). Visit the docs for more info.
P50 time to first token (TTFT) on live AI Gateway traffic, in milliseconds. View the docs for more info.
Direct request success rate on AI Gateway and per-provider. Visit the docs for more info.
More models by OpenAI
| Model |
|---|
| 1M | 1.8s | 62tps | $5.00/M | $30.00/M | Read: $0.5/M Write: — | $10.00/K + input costs | — | +4 |   | 04/24/2026 | |||
| 400K | 1.8s | 192tps | $0.75/M | $4.50/M | Read:$0.07/M Write:— | $10.00/K + input costs | — | +4 | | 03/17/2026 | |||
| 400K | 0.4s | 150tps | $0.20/M | $1.25/M | Read:$0.02/M Write:— | $10.00/K + input costs | — | +4 | | 03/17/2026 | |||
| 1.1M | 1.0s | 93tps | $2.50/M | $15.00/M | Read: $0.25/M Write: — | $10.00/K + input costs | — | +4 | | 03/05/2026 | |||
| 131K | 0.1s | 1403tps | $0.35/M | $0.75/M | Read:$0.25/M Write:— | — | — |   +5 | 08/05/2025 | ||||
| 1M | 0.4s | $0.10/M | $0.40/M | Read:$0.03/M Write:— | $14/K + input costs | — | +2 | | 04/14/2025 |
About GPT-5.3 Chat
GPT-5.3 Chat became available on March 3, 2026 on AI Gateway as a conversational variant of the GPT-5.3 generation. The update focuses on tone, relevance, and conversational flow, producing more accurate answers, better-contextualized web results, and fewer unnecessary refusals and caveats.
The model reduces hallucination rates compared to earlier chat variants and produces smoother, more direct responses. It supports the context window of 128K tokens, function calling, and multimodal input, providing the full feature set for chat applications.
If you're building chat-based products that rely on factual accuracy and natural conversation, GPT-5.3 Chat improves on previous generations in the areas that matter most for dialogue quality.
What To Consider When Choosing a Provider
- Configuration: GPT-5.3 Chat focuses on tone, relevance, and conversational flow. It produces more direct responses with fewer unnecessary caveats and refusals than previous chat variants.
- Configuration: The model reduces hallucination rates compared to earlier generations, making it more reliable for factual conversation.
- Zero Data Retention: AI Gateway does not currently support Zero Data Retention for this model. See the documentation for models that support ZDR.
- Authentication: AI Gateway authenticates requests using an API key or OIDC token. You do not need to manage provider credentials directly.
When to Use GPT-5.3 Chat
Best For
- Customer-facing chatbots: Natural, direct conversations with improved accuracy and fewer unnecessary refusals
- Web-augmented chat: Applications that surface contextualized web results within conversation
- Support automation: Multi-turn troubleshooting with smoother, more relevant responses
- Conversational search: Natural language exploration of topics with accurate, on-point answers
- Interactive assistants: Dialogue-based experiences where tone and conversational flow matter
Consider Alternatives When
- Non-conversational tasks: GPT-5.2 or GPT-5.4 for single-shot generation and structured processing
- Maximum capability: GPT-5.4 for the GPT-5.4 generation general-purpose model
- Cost-sensitive chat: GPT-5.4 mini or nano for more affordable conversational workloads
- Autonomous coding: GPT-5.3 codex for code-focused agent workflows
Conclusion
GPT-5.3 Chat advances conversational AI quality with improved accuracy, reduced hallucinations, and more natural dialogue flow. For chat applications on AI Gateway that prioritize reliable, direct responses, it's a strong option.
Frequently Asked Questions
How does GPT-5.3 Chat improve over GPT-5.2 chat?
It focuses on tone, relevance, and conversational flow. Expect more accurate answers, better-contextualized web results, fewer unnecessary refusals, and reduced hallucination rates.
What context window does GPT-5.3 Chat support?
128K tokens, supporting extended conversation histories.
Does GPT-5.3 Chat reduce hallucinations?
Yes. The model specifically targets hallucination rate reduction, producing more factually grounded responses than previous chat variants.
Is GPT-5.3 Chat suitable for non-chat tasks?
It can handle them, but GPT-5.2 or GPT-5.4 is generally better for single-shot generation, batch processing, and structured output tasks.
Does GPT-5.3 Chat support function calling?
Yes. It supports function calling, enabling conversational agents that interact with external APIs and tools during dialogue.
How does AI Gateway handle authentication for GPT-5.3 Chat?
AI Gateway accepts a single API key or OIDC token for all requests. You don't embed OpenAI credentials in your application; AI Gateway routes and authenticates on your behalf.
What are typical latency characteristics?
This page shows live throughput and time-to-first-token metrics measured across real AI Gateway traffic.