2 Sources
[1]
OpenAI GPT-6.2 Astra vs Fable 5.2: Use Case and Performance Comparison
OpenAI's GPT-6.2 Astra has just bee launched and has drawn mixed assessments, with benchmark costs and coding results presenting different pictures of its performance. David Ondrej compare high and extra-high reasoning settings and their benchmark results to identify a starting point for coding or
[2]
OpenAI's GPT-6.2 Astra Becomes The First Model To Achieve A 100% Score On ARC-AGI-3 While Being ~8x Cheaper Than Anthropic's Fable 5.2, But Actual Users Are Not Impressed
After a fairly long drought, which was presumably necessary to ensure proper alignment, OpenAI has finally launched its latest frontier-class model, dubbed GPT-6.2 Astra, to middling applause. OpenAI's GPT-6.2 Astra scores 62 on the Artificial Analysis Intelligence Index, while being 8 times
Share
Copy Link
OpenAI has launched GPT-6.2 Astra, a frontier-class model
1
2
. Per Wccftech, it is the first model to score 100% on ARC-AGI-3, a benchmark of interactive game-like environments from François Chollet's ARC Prize team, built to test fluid intelligence2
. User reactions in the first days have been mixed to negative2
.
Source: Geeky Gadgets
The benchmark win is the headline; the other charts are the problem. The model shows no improvement on the Artificial Analysis Coding Agent Index, matching GPT-6 Astra's score of 53
1
2
. Geeky Gadgets reads that as a slight gain on scores rounded to whole numbers1
. Vals AI ranks it seventh on its index, trailing several Anthropic and Google models, per Wccftech2
.The pricing is where OpenAI pushes hardest. Astra costs $8 per million input tokens and $40 per million output tokens, about 20% below its predecessor GPT-6 Astra
2
. On cost per task the gap is the pitch: about $4 per Artificial Analysis Intelligence Index task against Fable's $26.50, roughly 8x cheaper on Wccftech's math2
. Geeky Gadgets puts the gap at about seven times on a different benchmark1
. On intelligence the model still trails: Astra scores 62 on that index against Fable 5.2's 65, about 3 points behind (Geeky Gadgets' figures run 60.2 and 64.6)1
2
.Dan Shipper reported that his team, after weeks with the models, found the release disappointing next to Fable 5.2
1
. Astra often stopped too early on real coding tasks, wrote answers so brief they could be hard to follow, and showed good default design instincts with little variety1
. In one run it spent 30 minutes on a task and came back with a too-short answer instead of listing seven items1
. Users have reported it pausing long tasks to ask odd questions, producing output that could not be parsed, and getting stuck in loops2
. The criticism has sharpened with the wait, too; the models took a long time to arrive, and reviewers judged them against that wait1
.There are real wins. Per Wccftech, Astra is strong on science and knowledge benchmarks and leads in 3D and robotics
2
. It leads Agent's Last Exam at about $2 per task against Fable's $60, and its extra-high reasoning setting beats its max setting on some benchmarks1
. Beyond that the picture varies by test: Fable 5.2 leads Frontier Code and Gemini 4 Carbon tops Automation Bench1
. Astra's lead on LiveVal comes on a chart missing two rivals1
.Related Stories
Per the company's own rating, Astra is the first model OpenAI has publicly released rated 'High' for AI self-improvement
1
2
. Its system card presents the model as an early milestone on the road to fully automated self-improvement, and the model writes faster kernels for OpenAI's own chip1
2
. Full recursive self-improvement is still only expected by July 2027, per Vals AI1
. The two labs also sit partly opposite on protecting their models. OpenAI says Astra and its sibling Sol leak their underlying reasoning about 1% of the time2
, while Anthropic keeps its models off machine-learning tasks to discourage distillation1
. Per creator David Ondrej, Anthropic has surpassed OpenAI in revenue and valuation, a reversal of last year's positions1
.Per Tibo, an OpenAI figure, the divide came down to timing
2
. As Wccftech put it, a few people got early access a while back, and the team "then made a lot of changes in the" final 3 to 4 days without re-testing2
. The influencer videos were recorded before those changes landed2
. "Don't get me wrong, while the models are a hefty jump compared to their GPT predecessors, they are by no means perfect," he wrote2
. He added that the teams want the models to get better at the taste and niceness of use in future iterations, and that OpenAI hears the feedback2
.
Source: Wccftech
Summarized by
Navi
07 Sept 2026•Technology

02 Sept 2026•Technology

22 Sept 2026•Technology
