
Beli ARA Jual ARB | Registered Technical Analyst | Build @cuanivora
Joined November 2011
- Tweets12K
- Following859
- Followers1.5K
- Likes2.6K
Jujur ini bikin ekspektasi soal performa agak ambyar. Bawa bendera "6" ternyata yg dijual cost efficiency. Kenapa gak 5.7 aja ya kalo gini, atau jangan2 sdh pada ancang2 cloud agentic?😅
GPT-6 Sol and Luna push the cost efficiency frontier by halving cost relative to GPT-5.6 Sol and Luna. Intelligence Index and Coding Agent Index scores remain level with GPT-5.6, with progress in some evaluations and regressions in others
Pricing is approximately half that of GPT-5.6: Sol drops from $4/$20 to $2/$10 per million input/output tokens, and Luna from $0.20/$1.20 to $0.10/$0.50, with the same 90% discount for cache reads and 25% premium for cache writes.
Key takeaways:
➤ Halves Cost per Task: GPT-6 Sol (max) costs $1.06 per task to run the Artificial Analysis Intelligence Index, ~50% less than GPT-5.6 Sol (max) at $1.99. GPT-6 Luna (max) costs $0.07 per task, ~60% less than GPT-5.6 Luna (max) at $0.18. This is driven by the price cut, as both models use slightly more output tokens per task (31k vs 29k for Sol, and 51k vs 41k for Luna). These two releases allow OpenAI to capture a significant portion of the cost efficiency Pareto frontier.
➤ In the Coding Agent Index, Sol improves but Luna regresses: In OpenAI's Codex harness, GPT-6 Sol (max) scores 57 in the Artificial Analysis Coding Agent Index, up 2 points from GPT-5.6 Sol (max), with gains in Terminal-Bench 4.0 (43% vs 37%) and SWE-Atlas-QnA (58% vs 54%). At $2.99 per task it costs ~50% less than GPT-5.6 Sol (max) and sits on the Pareto frontier of Coding Agent Index vs Cost per Task. GPT-6 Luna (max) scores 41, down 2 points from GPT-5.6 Luna (max), with lower scores in SWE-Atlas-QnA (44% vs 49%) and DeepSWE v1.1 (64% vs 66%), at ~60% lower cost per task.
➤ Significant reduction in hallucination: Both models hallucinate less in AA-Omniscience, our knowledge and hallucination benchmark. GPT-6 Sol (max) cuts its hallucination rate from 92% to 60% and GPT-6 Luna (max) from 93% to 77%. Sol achieves this by declining to answer more often: it attempts 83% of questions vs 99% for GPT-5.6 Sol (max), which cuts wrong answers by about a quarter but also lowers accuracy 5 points from 59% to 54%. Luna's accuracy is broadly unchanged at 44% vs 43% while it answers fewer questions. On the AA-Omniscience Index, Sol improves from 22 to 27 and Luna from -10 to 1.
➤ Mix of improvement and regression across evals: Beyond AA-Omniscience, both models improve in AutomationBench-AA (Sol 62% vs 60%, Luna 53% vs 50%) and Terminal-Bench 4.0 (Sol 44% vs 40%, Luna 13% vs 12%). However, we observe regressions in two key knowledge work evaluations. In GDPval-AA v2.1, our benchmark adapted from OpenAI's dataset of economically valuable tasks across 44 occupations, Sol drops ~100 Elo points and Luna ~75. Luna also drops ~45 Elo points in AA-Briefcase v1.1, while Sol is level. AA-Briefcase v1.1 is a private evaluation across multi-week knowledge work projects, with thousands of input files. Our team has manually inspected hundreds of model outputs: the regressions tend to be driven by reduced presentation quality and deliverables that omit rubric elements.
Congratulations @OpenAI and @sama on the launch!
58
Sekali lagi, China kembali memenangkan perlombaan.
I agree with Dario that we need to pace the frontier. This has been a primary topic of discussions we've had at OpenAI in recent weeks.
Committing to having independent evaluators with employee-like access is a great idea, and we will do the same. We'll have more to share soon.
33
Soal seremonial segala hal, Indonesia tidak pernah beranjak kemanapun.
Tentara Jepang bingung. Datang ke Indo yg lagi kena bencana kebakaran hutan, tapi ternyata disambut meriah kayak lagi perayaan akbar 👏🏻
Kebakaran lahan hutan tu dalam itungan detik bakal terus meluas. Lebih penting seremonialnya, kerjanya tar dulu
Bangsa yg ramah, memang hobinya pesta 👏🏻👏🏻
67
Hidup di dunia apa kita ini..
#owetauinishitpost
you can't be serious
Readers added context they thought people might want to know
AI 🤖 generated image for Satire
Not to be confused as actual news
Tim Cook has not joined Anthropic; he is stepping down as Apple CEO to become executive chairman.
apple.com/newsroom/2026/…
x.com/tim_cook/statu…
9to5mac.com/2026/08/31/tim…
56
Kamu bisa mengikatnya di tiang pembangkit listrik. Biarkan dia menjadi generator listrik gratis sampai vaksin zombie ditemukan.
1
86
Bonus, presiden 2029 nanti asal Kediri.
Kediri:
Sekolah internasional ada.
Mall ada 3.
Bioskop ada.
RS jg banyak yg bagus2.
Gunung2 ada.
Jarak tempuh ke pantai 1,5 jam aja.
Bandara, stasiun, terminal bus ada semua.
Perumahan kluster banyak.
Tanah jg masih murah.
Tol bentar lagi nyambung ke trans Jawa.
UMR 2,3 jutaan.
103
Dari apa yg di flexingin, ketauan pejabat satu ini termasuk pejabat miskin. Pasti diketawain anak pejabat yg beneran kaya😅
Flexing dari anak Nevi Rizal, mantan PLT Sekda Gayo Lues.
Jembatan-jembatan di Gayo Lues belum diperbaiki dan bensin langka.
Meanwhile anak nya pamer di sosmed bahwa Nevi Rizal "nimbun minyak 15 jirigen" & pamer jalan2 ke luar negeri terus
Indah nya negeri Indonesia 🙂
105
Momen paling canggung sepanjang masa. Trump kasih trophy ke Spanyol😂
President Donald J. Trump presents the World Cup Trophy to Spain, the 2026 @FIFAWorldCup Champions! 🏆
61
Jelas aja dukung spain. Sengketa folkland aja masih jalan
Downing Street has given UK Prime Minister Keir Starmer's reaction to England's loss to Argentina in the World Cup semi-final.
Read more: bbc.in/3RaWkiF
113
Cicilan masih jalan 2 bulan, motor rusak gak berbentuk. Apes juga jadi WNI
Kepala Staf kepresidenan mengusulkan
Motor listrik yang udah terlanjur dipesan BGN, kedepannya bisa dicicil oleh pegawai SPPG agar dana APBN tidak terpakai
Okeh bertambah yang benci program MBG
Masyarakat ✅️
Investor SPPG ✅️
Pegawai SPPG ✅️
Ini program apasih ajg
34



































