GPT-6 Astra
No shutdown before this date. GPT-6 Astra stays on Amazon Bedrock until at least Sep 8, 2027.
≥ 11 mo- Earliest retirement
- ≥ 329 daysSep 8, 2027 · Amazon Bedrock
- Arena Text
- 1,475#22 of 128 · Arena
- GPQA Diamond
- 95.8%#2 of 108 · Epoch AI
- List price · $ per 1M
- $10.00 in $50.00 outOpenAI API · cached input $1.00
- Context window
- 1.1MMax output 128K tokens
Dates on every platform
| Platform | Model ID and status | Deprecated | Shuts down | Replacement |
|---|---|---|---|---|
| OpenAI APIOpenAI: List models (GET /v1/models) (opens in a new tab) · checked Oct 14, 2026 | gpt-6-astraActive | – | None set | – |
| Azure AI FoundryMicrosoft: Foundry Models sold by Azure (opens in a new tab) · checked Oct 14, 2026 | gpt-6-astra:2026-09-03Active | – | None set | – |
| Amazon BedrockAmazon Web Services: Amazon Bedrock ListFoundationModels, us-east-1 (opens in a new tab) · checked Oct 14, 2026 | openai.gpt-6-astraActive · also global.openai.gpt-6-astra, us.openai.gpt-6-astra | – | ≥ 11 moNot sooner than | – |
Available as Global Standard, US Data Zone Standard and Provisioned Throughput deployments.
In us-east-1, requests go through a cross-Region inference profile (listed in aliases).
What breaks when you move off GPT-6 Astra
- Parameters
- Parameters
- Tooling
- Behavior
Intelligence
| Benchmark | Publisher | Score | Rank | Published |
|---|---|---|---|---|
| Arena Text | Arena | 1,475 ± 3 | #22 of 128 | Oct 8, 2026 |
| GPQA Diamond | Epoch AI | 95.8% ± 1.4% | #2 of 108 | Aug 30, 2026 |
| Arena WebDev | Arena | 1,786 ± 6 | #2 of 69 | Oct 8, 2026 |
| Arena Agent | Arena | 0 ± 0 | #5 of 35 | Oct 12, 2026 |
| FrontierMath Tier 4 | Epoch AI | 97.6% ± 2.4% | #2 of 58 | Aug 30, 2026 |
| FrontierMath Tiers 1-3 | Epoch AI | 93.7% ± 1.4% | #1 of 71 | Aug 30, 2026 |
| OTIS Mock AIME 2024-2025 | Epoch AI | 100.0% ± 0.0% | #6 of 105 | Aug 30, 2026 |
| SimpleQA Verified | Epoch AI | 75.6% ± 1.4% | #2 of 68 | Aug 30, 2026 |
Arena (arena.ai), https://arena.ai/leaderboard
Epoch AI, ‘Capabilities & benchmarking’. Published online at epoch.ai. Retrieved from ‘https://epoch.ai/benchmarks’ [online resource].
List prices
| Platform | Input | Cached | Output | In effect | Source |
|---|---|---|---|---|---|
OpenAI APIgpt-6-astra | $10.00 | $1.00 | $50.00 | Oct 14, 2026Current · Requests over 272K input tokens are billed at higher long-context rates. Cache writes are billed separately. Batch and Flex bill less; Fast mode bills more. | OpenAI: Pricing (opens in a new tab) |
Azure AI Foundrygpt-6-astra:2026-09-03 | $10.00 | $1.00 | $50.00 | Sep 3, 2026Current · Global Standard short-context rate from the US East 2 price list. | Microsoft: Azure Retail Prices API (Foundry Models, US East 2) (opens in a new tab) |
Amazon Bedrockopenai.gpt-6-astra | $11.00 | $1.10 | $55.00 | Sep 1, 2026Current · Standard tier, in-Region or geographic cross-Region inference, which Bedrock prices 10% above global cross-Region inference. Long-context requests are billed at higher rates. | Amazon Web Services: AWS Price List API: AmazonBedrockFoundationModels, us-east-1 (opens in a new tab) |
Change history
ReleaseGPT-6 Astra available on Amazon BedrockAmazon Bedrock
Bedrock added GPT-6 Astra as openai.gpt-6-astra.
- Platform
- Amazon Bedrock
- Model
- GPT-6 Astra
Source: Amazon Web Services, Amazon Bedrock ListFoundationModels, us-east-1 (opens in a new tab)
ReleaseGPT-6 Astra available on Azure AI FoundryAzure AI Foundry
Microsoft made GPT-6 Astra generally available in Foundry with Standard and Provisioned Throughput deployments in Global and US Data Zone geographies.
- Platform
- Azure AI Foundry
- Model
- GPT-6 Astra
ReleaseGPT-6 Astra released on the OpenAI APIOpenAI API4 breaks
OpenAI released GPT-6 Astra for reasoning, coding, computer use, research and document creation, in the Responses and Chat Completions APIs.
- Platform
- OpenAI API
- Model
- GPT-6 Astra
- ParametersGPT-6 Astra does not support the none reasoning effort level.
- ParametersGPT-6 Astra does not accept custom temperature or top_p values, or log probabilities.
- ToolingTool calling requires the Responses API. Chat Completions requests with tools have to move to Responses.
- BehaviorMisalignment monitoring checks agent work in supported Responses API requests and can stop a conversation for review.