OpenAI model product
GPT-6-Astra
effort levels from low through max.
Updated Sep 8, 2026. Default version: GPT-6 Astra
LLMBoard Score98.9GPT-6 Astra
Coverage40%21 benchmark families
Context window1.1MTokens
Official input price$10OpenAI API
GPT-6-Astra Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
GPT-6-Astra Benchmark Results
Benchmark scores for GPT-6 Astra.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkAgents' Last Exam | Score59.30% | Rank01 | Participants16 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkARC-AGI | Score98.50% | Rank01 | Participants9 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkARC-AGI-3 | Score99.90% | Rank01 | Participants5 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkARC-AGI v2 | Score95.00% | Rank01 | Participants18 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkBenchCAD (with Python tool) | Score95.90% | Rank01 | Participants4 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkBrowseComp | Score91.50% | Rank01 | Participants63 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkExploitBench | Score100.00% | Rank01 | Participants6 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkExploitGym | Score42.40% | Rank01 | Participants5 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkFrontierMath Tier 4 (v2) | Score97.60% | Rank01 | Participants4 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkGeneBench-Pro | Score37.80% | Rank01 | Participants4 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkGPQA | Score96.00% | Rank01 | Participants247 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkLifeSciBench | Score60.30% | Rank01 | Participants4 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkScreenSpot Pro | Score92.70% | Rank01 | Participants26 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkSEC-bench Pro | Score85.40% | Rank01 | Participants4 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench 4.0 | Score57.70% | Rank01 | Participants13 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench-Science 0.1 | Score64.60% | Rank01 | Participants2 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkLM Arena Webdev | Score1,797.04 rating | Rank01 | Participants97 | Percentile100.00% | EvidenceA | Evaluated |
| BenchmarkDeepSWE 1.1 | Score74.10% | Rank02 | Participants31 | Percentile96.67% | EvidenceC | Evaluated |
| BenchmarkHealthBench Professional | Score63.40% | Rank02 | Participants10 | Percentile88.89% | EvidenceC | Evaluated |
| BenchmarkHumanity's Last Exam (with tools, text-only) | Score57.20% | Rank02 | Participants5 | Percentile75.00% | EvidenceC | Evaluated |
| BenchmarkOSWorld 2.0 | Score72.60% | Rank02 | Participants11 | Percentile90.00% | EvidenceC | Evaluated |
| BenchmarkFrontierCode 1.1 | Score53.30% | Rank03 | Participants17 | Percentile87.50% | EvidenceC | Evaluated |
| BenchmarkAutomationBench | Score41.40% | Rank04 | Participants18 | Percentile82.35% | EvidenceC | Evaluated |
GPT-6-Astra Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenawebdev | Categoryoverall | Rank01 | Rating / score1,797.04 | Votes1,199 | ObservationsN/A | Result date |
| Arenawebdev | Categorywebdev | Rank01 | Rating / score1,797.04 | Votes1,199 | ObservationsN/A | Result date |
| Arenawebdev | Categorywebdev-html | Rank01 | Rating / score1,724.83 | Votes149 | ObservationsN/A | Result date |
| Arenawebdev | Categorywebdev-react | Rank01 | Rating / score1,814.85 | Votes891 | ObservationsN/A | Result date |
GPT-6-Astra Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $10 input, $50 output per 1M
- Official provider
- OpenAI
- Lowest third-party
- From $10 input, $50 output per 1M via NanoGPT
- Tracked offerings
- 18
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderNanoGPT | Provider model IDopenai/gpt-6-astra | Regionglobal | Input / 1M$10 | Output / 1M$50 | Context1.1M | Updated |
| ProviderLLM Gateway | Provider model IDopenai/gpt-6-astra | Regionglobal | Input / 1M$10 | Output / 1M$50 | Context1.1M | Updated |
| ProviderCrossModel | Provider model IDopenai/gpt-6-astra | Regionglobal | Input / 1M$10 | Output / 1M$50 | Context1.1M | Updated |
| ProviderVercel AI Gateway | Provider model IDopenai/gpt-6-astra | Regionglobal | Input / 1M$10 | Output / 1M$50 | Context1.1M | Updated |
| ProviderVenice AI | Provider model IDopenai-gpt-6-astra | Regionglobal | Input / 1M$10 | Output / 1M$50 | Context1.1M | Updated |
| ProviderDevPass (LLM Gateway) | Provider model IDgpt-6-astra | Regionglobal | Input / 1M$10 | Output / 1M$50 | Context1.1M | Updated |
| ProviderGitHub Copilot | Provider model IDgpt-6-astra | Regionglobal | Input / 1M$10 | Output / 1M$50 | Context1.1M | Updated |
| ProviderKilo Gateway | Provider model IDopenai/gpt-6-astra | Regionglobal | Input / 1M$10 | Output / 1M$50 | Context1.1M | Updated |
| ProviderMerge Gateway | Provider model IDopenai/gpt-6-astra | Regionglobal | Input / 1M$10 | Output / 1M$50 | Context1.1M | Updated |
| ProviderOpenCode Zen | Provider model IDgpt-6-astra | Regionglobal | Input / 1M$10 | Output / 1M$50 | Context1.1M | Updated |
| ProviderOpenRouter | Provider model IDopenai/gpt-6-astra | Regionglobal | Input / 1M$10 | Output / 1M$50 | Context1.1M | Updated |
| ProviderOpenAI | Provider model IDgpt-6-astra | Regionglobal | Input / 1M$10 | Output / 1M$50 | Context1.1M | Updated |
| ProviderEden AI | Provider model IDopenai/gpt-6-astra | Regionglobal | Input / 1M$10 | Output / 1M$50 | Context1.1M | Updated |
| ProviderAzure | Provider model IDgpt-6-astra | Regionglobal | Input / 1M$10 | Output / 1M$50 | Context1.1M | Updated |
| ProviderOfox | Provider model IDopenai/gpt-6-astra | Regionglobal | Input / 1M$10 | Output / 1M$50 | Context1.1M | Updated |
| ProviderNeon | Provider model IDgpt-6-astra | Regionglobal | Input / 1M$10 | Output / 1M$50 | Context1.1M | Updated |
| ProviderVivgrid | Provider model IDgpt-6-astra | Regionglobal | Input / 1M$10 | Output / 1M$50 | Context1.1M | Updated |
| ProviderVercel AI Gateway | Provider model IDopenai/gpt-6-astra-fast | Regionglobal | Input / 1M$20 | Output / 1M$100 | Context1.1M | Updated |
GPT-6-Astra Runtime Performance
Provider-specific output speed and catalog latency for GPT-6 Astra. Runtime does not affect the capability score.
Sort by
Provider | Output Speed | Catalog Latency | Max Input | Max Output | Updated |
|---|
| ProviderOpenAI | Output Speed30.66 tok/s | Catalog Latency14.00 s | Max Input1.1M | Max Output128K | Updated |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
GPT-6-Astra Specifications
Technical details for the model's default version.
- Version
- GPT-6 Astra
- Released
- Sep 4, 2026
- Knowledge cutoff
- Apr 30, 2026
- Parameters
- N/A
- Context window
- 1.1M
- Max output
- 128K
- Inputs
- image, text
- Outputs
- text
- Open weights
- No
- License
- Proprietary
GPT-6-Astra Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionGPT-6 Astra | Released | LLMBoard98.89 | ParametersN/A | Context1.1M | Max output128K | Open weightsNo | LicenseProprietary |
Models similar to GPT-6-Astra
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
OP
GPT-5.6-Sol
OpenAI
OP
GPT-5.6-Terra
OpenAI
OP
GPT-5.5
OpenAI
OP
GPT-5.4
OpenAI
OP
GPT-5.5-Pro
OpenAI
OP
GPT-5.6-Luna
OpenAI
What is GPT-6-Astra?
Key information about GPT-6-Astra and its available data.
GPT-6-Astra is an OpenAI large language model for reasoning, coding, computer use, research, and document creation. 05M-token context window and a maximum output of 128K tokens.
Data as of 2026-09-08.
FAQ
Common questions about GPT-6-Astra.
When was GPT-6-Astra released?
GPT-6-Astra's default version was released on Sep 4, 2026.
How much does GPT-6-Astra cost?
GPT-6-Astra's official API price is $10 per million input tokens and $50 per million output tokens via OpenAI. The lowest tracked third-party offer starts at $10 input and $50 output via NanoGPT.
Who created GPT-6-Astra?
GPT-6-Astra was created by OpenAI.
What is the context window for GPT-6-Astra?
The default version has a 1.1M token context window.
Is GPT-6-Astra open weight?
No. The default version is not marked as having publicly available weights.
How many API providers offer GPT-6-Astra?
18 provider offerings are linked to the default version.