Skip to main content

GPT-6 Astra

GPT-6 Astra is OpenAI's flagship model for demanding end-to-end work.

Input / output priceUp to 18% off
$10.00 / $50.00$8.20 / $41.00 per 1M
Context
1.1M tokens
Released
Sep 3, 2026
Agent setup

Paste into Claude Code, Cursor, or Codex

Help me set up the Respan gateway: fetch and follow https://www.respan.ai/docs/agent-setup.md. Use model "openai/gpt-6-astra".

Read docs

Providers

9 routes from 7 providers, cheapest first, retired then OpenRouter last, priced per 1M tokens after gateway discounts.

ProviderInputOutput
OpenAI18% offopenai/gpt-6-astra
$10.00/M$8.20/M$50.00/M$41.00/M
Azure Foundryazure_foundry/gpt-6-astra
$10.00/M$50.00/M
Azure OpenAIazure_openai/gpt-6-astra
$10.00/M$50.00/M
Bedrockbedrock/global.openai.gpt-6-astra
$10.00/M$50.00/M
Perplexityperplexity/gpt-6-astra
$10.00/M$50.00/M

Is a provider down? ChatGPT status, Azure OpenAI status, AWS Bedrock status, Perplexity status, OpenRouter status

Pricing

Averaged across 9 routes from 7 providers, in USD per 1M tokens after gateway discounts.

Average input
$9.36 per 1M
Average output
$46.78 per 1M
Average cache read
$0.94 per 1M
Average cache write
$11.69 per 1M

Cost by volume

Total cost from 0 to 1T tokens on the cheapest and most expensive provider.

Highest$11.00/MBedrockLowest$5.00/MOpenRouter

Throughput and latency

30-day averages from requests through the Respan gateway. Under each figure, every model on the gateway by its average, this one in blue.

Output speed

Higher is better

62tokens/sacross providers

Faster than 70% of the other 596 models on the gateway

Time to first token

Lower is better

2.8sacross providers

Faster than 35% of the other 153 models on the gateway

Response latency

Lower is better

17sacross providers

Faster than 11% of the other 776 models on the gateway

Benchmarks

The Respan Index plus 34 published scores from 14 sources, each linked to where it was published.

176 ranked models. The shaded band is this model's 95% interval, 68.6 to 75.7. Claude Sonnet 5.5 and Claude Fable 5.1 fall inside it, so the gap is within the margin of error.
GPT-6 Astra by category: Coding 58.5, Agents 78.3, Reasoning 86.0, Knowledge & language 75.6, Human preference 60.1
  1. 7GPT-6 Astra72.3
  2. 6Claude Sonnet 5.573.4
  3. 7Claude Fable 5.172.3

Published scores

BenchmarkScore
WeirdMLHåvard Tveit Ihle93.6%
LiveBench CodingLiveBench80.4%
GSO Opt@1GSO77.5%
DeepSWE v1.1Datacurve74.1%
FrontierSWE V2Proximal Labs65.5%
Terminal-Bench Science 0.1OpenAI64.6%
MirrorCodeEpoch AI60%
Terminal-Bench 4.0Vals AI59.6%

Compare GPT-6 Astra: GPT-6 Astra vs GPT-6.2 Astra, GPT-6 Astra vs Gemini 4 Argon, GPT-6 Astra vs Phi-4, GPT-6 Astra vs Claude Sonnet 5.5, GPT-6 Astra vs GPT-5, GPT-6 Astra vs GPT-6.1 Sol

Apps

Use GPT-6 Astra in the tools you already run. Each guide points the app at the Respan gateway.

Setup guides for more apps are in the integration docs.

GPT-6 Astra FAQ

Short answers, built from the data on this page.

GPT-6 Astra is a model from OpenAI, released on Sep 3, 2026. GPT-6 Astra is OpenAI's flagship model for demanding end-to-end work.

On the Respan gateway, GPT-6 Astra costs from $5.00 per 1M input tokens and $25.00 per 1M output tokens on OpenRouter, after gateway discounts. Averaged across its 9 routes from 7 providers, it costs $9.36 per 1M input tokens and $46.78 per 1M output tokens.

GPT-6 Astra is available through 9 routes from 7 providers: OpenAI, Azure Foundry, Azure OpenAI, Bedrock, Perplexity, Inference.net, and 1 more.

GPT-6 Astra accepts up to 1.1M tokens of context on its largest provider route.

Point any OpenAI-compatible SDK at https://gateway.respan.ai/v1 with your Respan API key and set the model to "openai/gpt-6-astra". To set it up with a coding agent, copy the agent setup prompt at the top of this page.

GPT-6 Astra scores 72.3 out of 100 on the Respan Index v1.1, ranking #7 of 176 models. By category: Coding 58.5, Agents 78.3, Reasoning 86.0, Knowledge & language 75.6, and Human preference 60.1. The Respan Index combines independent and lab-reported benchmark results into one score, using benchmark data collected by Respan.

GPT-6 Astra has 34 published benchmark scores, including WeirdML 93.6%, BALROG 68.3%, and GPQA Diamond 96%. Each score links to where it was published; the data is collected by Respan.

Certified where it counts.

Built to meet the security and privacy standards that enterprise and healthcare teams require.

ISO 27001

The internationally recognized standard for information security management.

SOC 2

Secure, compliant management of your data across all of our systems.

GDPR

Operated under GDPR, the world's strictest standard for data privacy.

HIPAA

HIPAA compliant, with a BAA available for healthcare teams.

Works with your entire stack

Use Respan with your favorite frameworks and tools.

AnthropicAI21 LabsxAIVercel AI SDK
LangChainOpenRouterAssemblyAICursor
LiteLLMGroqBasetenMem0
CohereCogneeLlamaIndexFireworks
DeepSeekLinkupHaystackTogether AI
PerplexityInference.netPostHogRubyLLM
SlackMastraAzure OpenAIMistral
AWS BedrockMoonshotHyperspellBAML
ModaInstructorGoogle GeminiParasail
ReductoZapierAgnoNebius
OpenAINovita AIReplicateSuperagent

Built for AI agents.
Break less.
Ship more.