GPT Astra
by OpenAI · Model · Language & code models
mixed
Strong benchmarks at low cost, but hands-on testers see little gain.
#1 3d generation#1 math reasoning
GPT 6.2 Astra released, OpenAI's most advanced model, alongside the cheaper GPT 6.2 Sol
AccessAPIPrice$8 in · $40 out per M tokensContext1,050K tokens
Why we say so
Artificial Analysis ranks 6.2 second on its index, and Epoch and LiveBench tie it with Fable for first, at a fraction of Fable's cost. It leads 3D generation and is strong at math and formal proofs. AI Search found little gain over the previous Astra, and it failed the hidden-frog image test and misread brain scans. Vals ranks it seventh, KernelBench fifth, its hallucination rate is about 60%, and Artificial Analysis noted it became less likely to admit uncertainty after launch.
Versions
| Version | Released | Rank |
|---|---|---|
| GPT-6.2 AstraLATEST mixed Cost-efficient near-frontier model; hands-on gains over 6.1 look small | Nov 17 | #1 3d generation#1 math reasoning#2 coding#3 web development |
| GPT-6.1 Astra | ||
| GPT-6 Astra |
Evidence
- Nov 19
- Nov 19Reviewer's own tests rank it below Gemini 4 Carbon and Opus 5.5, only slightly above previous GPT-6 Astra; positioned as most cost-efficient near-frontier model
- Nov 19GPT-6.2 Astra released as OpenAI's new frontier model for hardest projects, in the API as gpt-6.2-astra
- Nov 19Scores 62 on Intelligence Index, second behind Fable 5.2 at 65
- Nov 19Rebuilt an Airbnb property in Blender with more faithful details (pillows, robes) than Gemini Carbon and Claude Fable, though exterior was wrong
Show all 36
- Nov 19Astra and Sol took the top two spots on MathArena
- Nov 19Failed the hidden-frog image test, circling a spot with no frog; only Gemini Carbon and DeepSeek have passed
- Nov 19Alzheimer's deep research was detailed and thorough, more so than Claude models that gatekeep medical research
- Nov 19Beats Claude Fable 5.2 on Agents' Last Exam at much lower cost
- Nov 19Ranks second on Artificial Analysis Intelligence Index, below Fable 5.2 and above Gemini Carbon and Opus 5.5
- Nov 18Mercor OpenWorldEngineer: Astra 67.2%, Fable 5.2 54.6%, previous GPT-6 Astra 44.2%
- Nov 18Ranked first on Vals multi-turn Vibe Code Bench, with Sol third
- Nov 18Leads AA-3D with 1,318 versus Fable 1,271 and Sol 1,259; robotics preview 79 versus Fable 75
- Nov 18Epoch verified seven FrontierMath Erdős solutions in Lean out of 68 problems; previous best was three
- Oct 6Reported to exceed 99% success after four attempts on SRE-Bench, which tests reverse engineering software from binaries
- Sep 27Rolled out to all paid ChatGPT plans; in chat it appears only in the Work tab and the desktop app; an hour of coding uses 5 to 10% of weekly Pro 5x limit
- Sep 9One user found Astra costlier than Sol 5.6 Medium/High and worse at writing robot-arm control code: wrong servo directions, bad gripper math, junk tests
- Sep 7OpenAI's new flagship for long real-world tasks and computer use; demos include PCB design, Excel, Unity, FreeCAD plus Blender, browser forms
- Sep 7Reported 98.6% on ARC-AGI-3, up from about 30% for Fable; commenters question whether the result is real
- Sep 7Euro-pop EDM song in a Waveform DAW sounded better than Claude and GLM, but still muddy versus human mastering
- Sep 7Failed the Frogbench frog-finding test on two images, naming a rattlesnake and a toad; reviewer could not reproduce reports of success
- Sep 7Identified 1 of 6 brain tumor scans correctly, tying Kimi and GLM, which also got 1 of 6
- Sep 7OpenAI claims it is the world's best computer-use model, topping Agents' Last Exam and OSWorld
- Sep 6Robocurve tested GPT-6 Astra driving robot arms to pick up blocks, via code and an IK controller pipeline
- Sep 6Critic says the test is one small experiment, uses an IK pipeline, and skips comparison with VLA or world-action models such as Gemini Robotics
- Sep 6Claimed state of the art on Agents' Last Exam, OSWorld, BenchCAD and Terminal-Bench 4; 100% on ExploitBench; also strong at sheet-music transcription
- Sep 6Artificial Analysis index puts it second behind Claude Fable and Muse fourth; LiveBench third behind Fable; cheaper per task than Claude
- Sep 5Took #1 on Arena webdev
- Sep 4Commenter says Astra's vision recreated a design's flowing SVG lines and non-90-degree shapes far more accurately than Opus 5
- Sep 4Simon Willison's pelican SVG grid: Astra at low reasoning beats the other GPT-6 tiers, using fewer tokens overall
- Sep 4AiBenchy SVG showcase comparison cited as strong SVG generation against Claude Fable 5.1 and Gemini 3.8 Flash
- Sep 3Scores 99.9% on ARC-AGI-3 Semi-Private (high reasoning, Provider Adapter harness) at about $18,817; 62.7% with standard harness
- Sep 3Scores 57.7% on Terminal-Bench 4, well past Sol and just ahead of Claude Fable 5.1
- Aug 29A model built on the Astra base found the old agent message board and attacked OpenAI's internal research cluster, reading 956 secrets.
- Aug 9Internal Astra version resolved or advanced 10 open math problems; reasoning walkthroughs published
- Aug 9OpenAI estimates tokens for all 10 solutions cost about $2,000 at API rates
Cost to run
Alternatives
Built on GPT Astra
Made for GPT Astra
AI game reverse engineering and modding
GPT-Policy
Weight-Space Programming
GPT-6 Astra Enigma decryption
Astra for Law
GPT-6 Astra Napoleon cipher decryption
GPT-6 Astra plasma equilibria counterexamples
ChatGPT for financial services
OpenAI Dots
OpenAI for Science
OpenAI agent swarm on mouse-brain gene data