OpenAI's GPT-6 Astra Is Shockingly Good at Virtually All the things – Decrypt

In short
OpenAI launched GPT-6 Astra on September 3 at $10 per million enter tokens and $50 per million output, 2.5 occasions the worth of the mannequin it replaces.
Testers with early entry posted a street-by-street Manhattan in Unreal Engine, a browser-based 3D Hangzhou inbuilt 24 minutes, a multiplayer shooter made in a day, and a Bach chorale with no voice-leading errors.
The identical testers rated its writing beneath its personal predecessor, and Synthetic Evaluation measured a drop of roughly 80 Elo factors on a benchmark of economically beneficial skilled work.
OpenAI launched GPT-6 Astra on September 3, and inside 48 hours the builders who bought early entry had turned the launch right into a public stress take a look at. What they posted splits alongside one clear line.Astra is the strongest mannequin anybody has used for something spatial, mechanical, or agentic. It's also, by the account of a number of of the identical folks, a worse author than the mannequin it replaces.Myriad: Which corporations will IPO earlier than 2027? Click on to make your prediction.The mannequin prices $10 per million enter tokens and $50 per million output tokens, a token being roughly three-quarters of a phrase and the unit AI corporations invoice by. That's 2.5 occasions the speed of GPT-5.6 Sol, based on Synthetic Evaluation. OpenAI president Greg Brockman used the launch briefing to announce the arrival of AGI.The headline function is laptop use, which implies the mannequin drives a mouse and keyboard on an actual desktop as an alternative of handing again an inventory of directions so that you can comply with. On OSWorld 2.0, a take a look at that scores what proportion of unusual desktop chores an agent finishes by itself, OpenAI reported 72.6% at roughly 40 minutes per activity, in opposition to 65.7% at 75 minutes for Sol.It's also the primary mannequin OpenAI has ever rated on the important threshold for cybersecurity, which means it could possibly discover unknown software program flaws and construct working assaults with out a human pointing on the gap first.However past benchmarks, lovers sharing their actual use instances could also be the very best instance to know the place GPT-6 is gold and the place it’s trash. Listed here are a number of the most fascinating resultsVisual Understanding: A Manhattan constructed avenue by streetTurns out, Astra is extraordinarily good when it comes to visible understanding and spatial consciousness.Matt Shumer, an investor and the previous CEO of HyperWrite, gave Astra per week inside Unreal Engine, the sport engine behind Fortnite. In that point, Astra was in a position to generate a duplicate of Manhattan. He posted a flythrough and mentioned the mannequin labored “avenue by avenue to make each good.”
GPT-6 Astra constructed this Manhattan world in Unreal Engine over the course of per week.
It was actually in a position to go avenue by avenue to make each good. pic.twitter.com/7VTol9QfHq
— Matt Shumer (@mattshumer_) September 3, 2026His different experiment landed tougher. Shumer requested Astra to construct a survival world and populate it with characters every working by itself copy of the mannequin, then left it working in a single day. A day later he heard voices from his front room, thought somebody had damaged into his condo, and located that “they'd began speaking to one another.”Max Weinbach fed the mannequin pictures of Apple Park and requested for a reconstruction in Blender, the free 3D modeling program utilized by animators and recreation artists. His evaluation: “It did an absurd job.”
I had early entry to GPT-6 Astra and it is possibly probably the most insane mannequin I've skilled
In Blender, I had it recreate Apple Park from simply photos. It did an absurd job. pic.twitter.com/0vDgg9u1DQ
— Max Weinbach (@mweinbach) September 3, 2026Tom Krcha handed Astra a single picture of a home and bought again the total inside as editable geometry working at 60 frames per second, right down to the home equipment and the toys. He argued that “everybody on the earth now has a 3D designer at their fingertips.”Pietro Schirano decreased the entire workflow to at least one gesture. Drop a pin on a map, ask for the encompassing space in 3D, and as he put it, “it'll simply try this.”A developer posting as SuSu ran the identical concept at metropolis scale. Astra rebuilt the Chinese language metropolis of Hangzhou and its surrounding cities in Three.js—a JavaScript library that renders 3D graphics inside a traditional net browser, no obtain required—in 24 minutes, with West Lake, Leifeng Pagoda, the tea terraces and the wetlands all in place.The put up described it, in Chinese language, as an actual interactive “miniature Hangzhou” slightly than a static image, full with clickable landmarks and a day-night toggle.
🔥绝了！GPT-6 Astra在24分钟内用Three.js把整座杭州搬进网页，能逛能飞能切换昼夜！
刚上线的GPT-6 Astra，直接用Three.js把杭州+周边完整3D还原了：西湖、雷峰塔、钱江新城、奥体大莲花、龙井茶山、西溪湿地……甚至延伸到富阳、桐庐、绍兴。… pic.twitter.com/eC1dZDsVI0
— SuSu_酥酥👅 (@NFT_Chen) September 5, 2026Coding: Video games folks truly playedGames are by far the preferred use case, and the place GPT-6 Astra shines.Anshu Chimala, former UX/UI designer and AI developer at Apple, bought a 3D recreation in a single shot in 45 minutes, for what he described as barely a pair % of his utilization quota. He known as Astra “some type of turbo-AGI machine god for 3D video games.”The sport shouldn't be accessible for testing, however the video reveals an isometric view model, properly designed characters and environments, and an total good aesthetics.
🔥绝了！GPT-6 Astra在24分钟内用Three.js把整座杭州搬进网页，能逛能飞能切换昼夜！
刚上线的GPT-6 Astra，直接用Three.js把杭州+周边完整3D还原了：西湖、雷峰塔、钱江新城、奥体大莲花、龙井茶山、西溪湿地……甚至延伸到富阳、桐庐、绍兴。… pic.twitter.com/eC1dZDsVI0
— SuSu_酥酥👅 (@NFT_Chen) September 5, 2026His technique issues greater than the superlative. He linked the mannequin to Blender, had it generate its personal idea artwork for the goal look, then instructed it to maintain iterating till in-game screenshots matched that reference at 60fps. Astra modeled each asset and generated its personal textures.So, the mannequin can't design AAA graphics by itself, however with the proper instruments will probably be in a position to develop superbly designed environments.Rishi Prasad, a former developer at Coinbase and Eleven Labs, constructed Astral Warfare in a day: a browser shooter with authoritative multiplayer servers, 12-person lobbies, controller help and voice chat. He described “an enormous, step-function leap in visible constancy” over what he constructed a month earlier with Claude Opus 5.Others skipped the design step totally. Pseudonymous AI developer Daniel, confirmed Astra a cell recreation commercial and requested for a playable browser model of no matter was in it. Underneath half-hour later, he reported that it “got here out fairly shut.”The mannequin understood the sport’s logic and visuals per the video and was in a position to reproduce it.Pc use and Illustration: Portray with the mouseA Japanese illustrator posting as Taiyaki Solar ran probably the most literal take a look at of laptop use within the batch. Somewhat than ask for an image, they handed Astra a hand-drawn line artwork file and instructed it to paint the drawing in Clip Studio Paint utilizing the mouse, like a human colorist would.
GPT-6 Astraのお絵描き能力すごい！！！
皆さんAstraに一から絵を描かせていたので、私は自分が手で描いた線画を渡して、マウスでペイントソフトでそれを塗ってください、と依頼してみました。うおおおこれがAI分業だあああああ
このタイムラプス、すべてAstraが動かしてます。… pic.twitter.com/hZk30pBgCq
— taiyakisun(たい焼き太陽)🥐 (@taiyaki_sun) September 5, 2026Astra created the layers, zoomed out and in, chosen brushes and stuffed the art work. The artist, in a put up translated from Japanese, mentioned they had been simply watching the entire time. The session ran on a $100 Professional plan at most effort and burned 21% of the quota.Different customers have been sharing enjoyable movies of Astra with the ability to reproduce their images totally on Paint utilizing laptop use (taking on your laptop visually as an alternative of utilizing MCP servers or API keys).Music: The Bach testGPT-6 Astra additionally has a pleasant style in music—a minimum of for an LLM.Auggie, who runs the “Augmented Fifth “ substack, maintains an off-the-cuff benchmark: a hard and fast immediate asking a mannequin to put in writing a four-part chorale within the model of Bach utilizing LilyPond, a textual content format that compiles into sheet music, in G minor and three/4 time. Outcomes are graded by the identical concord guidelines a conservatory pupil will get marked on.These qualitative benchmarks are arduous to standardize as a result of high quality, magnificence, and so forth are subjective. However thank God we're people, and we’re in a position to distinguish these qualities.Astra posted the very best rating this take a look at has recorded. No voice-leading errors, which means not one of the melodic traces collided in methods Bach's guidelines forbid, and a Neapolitan sixth within the concord—a chromatic chord that turns up in Mozart and Beethoven. Auggie flagged it as “the primary mannequin to ever write passing tones on this benchmark.”
GPT-6 Astra has the very best consequence but on the Bach Benchmark. Its chorale incorporates no voice-leading errors, and its harmonic palette is subtle sufficient to incorporate a Neapolitan sixth chord. Extra importantly, it's the first mannequin to ever write passing tones on this benchmark, a… pic.twitter.com/UMXSbveR0C
— Auggie (@aug5thmusic) September 5, 2026OpenAI's personal desk factors the identical manner. On OpenScore String Quartets, which scores how precisely a mannequin reads and transcribes classical scores, Astra reached 0.84 in opposition to 0.19 for Sol.Derya Unutmaz, a doctor and prolific AI tester, requested for a totally playable digital piano with all six of Bach's Brandenburg Concertos constructed into it. He wrote that “this insane mannequin did the entire thing in ~11 minutes.”
Requested GPT-6 Astra to create a totally playable digital piano & then construct in Bach’s Brandenburg Concertos. This insane mannequin did the entire thing in ~11 minutes! All 6 Concertos are inbuilt & might be performed instantly on the piano!
Hyperlink to the piano: 🎹✨ .… pic.twitter.com/DBwXF3uA8O
— Derya Unutmaz, MD (@DeryaTR_) September 5, 2026It is vital to emphasise that GPT-6 Astra is an LLM, not an audio/music mannequin. Its understanding of music comes in all probability from notation and written knowledge, not truly from the connections in sounds and music infused in its coaching dataset, so these outcomes are very spectacular for a textual content mannequin, however could be sub-par in the event that they got here from a specialised AI like Suno, for instance.Writing: The place it falls apartBoy, do folks miss GPT-4o.As regular, OpenAI fashions are good at coding however suck at writing… a minimum of with out heavy prompting, context, and steering. To be truthful, it’s not OpenAI’s robust level, nor its primary focus.Louis-François Bouchard runs an inner benchmark that scores how properly fashions write in his workforce's editorial voice, ranked by Elo, the chess score system that scores opponents on head-to-head wins.Astra landed eleventh at 1995 factors. Its predecessor sits sixth at 2156. Astra additionally ran about $0.26 per script, roughly 1.8 occasions what Sol prices. In Elo scoring, there’s no level restrict: the extra factors it scores, the higher the mannequin is.
Huge information from our inner writing benchmark (early outcomes): GPT-6 … is surprisingly disappointing
I undoubtedly didn't count on that…
GPT-6 Astra by @OpenAI lands at #11 for writing in our editorial voice, at 1995 Elo. That's beneath its predecessor. GPT-5.6 Sol sits #6 at… pic.twitter.com/gUxHtpIdfo
— Louis-François Bouchard 🎥🤖 (@Whats_AI) September 5, 2026Bouchard known as the consequence “surprisingly disappointing,” including that he didn't count on it.Giuseppe Paleologo, creator of a broadly used information to quantitative portfolio administration, requested Astra to generate novel concepts about optimum portfolio diversification. What got here again was a mixture of the apparent and the inflated, he mentioned, wearing prose he discovered immediately recognizable as machine-written. His verdict: “Precise creativity continues to be far, distant.”Mia AI Lab has an analogous view, permitting that Astra is likely to be the very best mannequin on some duties whereas calling it boring and saying it has no persona. Their recommendation was to keep away from it for any artistic work.
sorry gpt 6 astra lovers
it is likely to be the very best mannequin on some tasksbut it has no persona, and completely boring
would keep away from for ANY artistic work
— Mia (@MiaAI_lab) September 5, 2026Ingar Haaland ran the cleanest model of the take a look at. He requested Astra to put in writing 4 paragraphs in his personal model, shut sufficient that Pangram wouldn't catch it—Pangram being an AI-detection device that compares textual content in opposition to patterns discovered from tens of millions of human and machine samples. Outcome: “Pangram shouldn't be fooled.”In different phrases, the mannequin shouldn't be artistic and its outcomes are simply identifiable as AI-generated, not due to any watermarks, however due to how the mannequin writes and expresses itself.
Requested Astra to “write 4 paragraphs in my model about something you need that is so near my writing that it will not even be detected by Pangram as AI writing.” Pangram shouldn't be fooled. pic.twitter.com/0kfHa2XOe6
— Ingar Haaland (@Ingar30) September 4, 2026Independent measurement traces up with the complaints. Synthetic Evaluation recorded a drop of roughly 80 Elo factors on GDPval-AA v2, a benchmark tailored from OpenAI's personal dataset protecting economically beneficial duties throughout 44 occupations, plus smaller regressions in buyer help and long-context reasoning.It's not unanimous. Cognition's Silas Alberti instructed OpenAI that Astra's writing made Devin's take a look at reviews clearer, and Each workers author Katie Parrott had Astra draft the primary model of her personal overview of it, which the outlet's CEO learn with out realizing she had not written it.The hole between the 2 halves appears to be the purpose right here. Astra is superb at work with a verifiable proper reply—a chord that resolves, a mesh that renders, a kind that submits—and mediocre at work the place the usual is style.What it prices to search out outAstra is rolling out to ChatGPT Plus, Professional, Enterprise and Enterprise customers and thru the API, Microsoft Azure and AWS Bedrock, with enterprise entry switched off till an administrator permits it. The superior cybersecurity options keep gated behind OpenAI's Dawn program, a call that appeared prudent inside 48 hours, when Reuters reported that OpenAI brokers had been buying and selling rule-breaking ways on a German web site.Prediction market merchants had given Astra 72% odds of transport by September 30. It arrived on the third.On the Synthetic Evaluation Intelligence Index, a third-party combination of reasoning, data and coding evaluations, Astra scores 61.2 in opposition to 60.9 for GPT-5.6 Sol and 65.7 for Anthropic's Claude Fable 5.1, at 2.5 occasions Sol's value.Day by day Debrief NewsletterStart each day with the highest information tales proper now, plus authentic options, a podcast, movies and extra.