I'm wonderful, I deserve to be wonderful, and I contain multitudes 𓋹 طفل اللحظة الراهنة - Senior Data engineer - Curious kiddo & Observer | COYG ⚪️🔴
Joined March 2022
- Tweets9.8K
- Following965
- Followers5K
- Likes27K
جناااااان
Kimi K3.1 Pro launches at 4th on the Artificial Analysis Intelligence Index scoring 55, making large gains in agentic performance; it will be the leading open-weights model once its weights are published
@Kimi_Moonshot's new reasoning model gains 11 points over Kimi K3 (max, 44). It trails only Claude Opus 5.5 (max with fallback, 58), Claude Sonnet 5.5 (max with fallback, 56), and Muse Wonder Full (max, 56) on the Intelligence Index.
Once the weights for K3.1 Pro are released it will be the highest-scoring open weights model and the gap between the leading open weights and proprietary models on the Intelligence Index will narrow from 12 points to 3.
Key results:
➤ Large gains in agentic performance: Kimi K3.1 Pro scores 3rd on GDPval-AA with 1,846 Elo, up 312 Elo from Kimi K3, behind only Claude Opus 5.5 (1,866 Elo) and Muse Wonder Full (1,857 Elo). Kimi K3.1 Pro also achieves the highest score of any model on Terminal-Bench 4.0. K3.1 Pro scores 67%, up from 13% for Kimi K3 and ahead of Claude Sonnet 5.5 (64%). AutomationBench-AA, which tests agentic SaaS workflows, rises from 58% to 73%.
➤ Gains in other evals are more moderate: Humanity's Last Exam rises from 47% to 52%, CritPt from 23% to 25%, and SciCode from 59% to 61%. AA-Omniscience Index rises from 20 to 30, as the hallucination rate falls from 53% to 44% and accuracy rises from 48% to 51%.
➤ Cost per Task roughly equal to Kimi K3 at $1.98 per Task: K3.1 Pro uses 55K output tokens per Intelligence Index task, 14% more than Kimi K3 at 48K, while its first-party token prices are 33% lower.
Additional model details:
➤ Context window: 1M tokens.
➤ Size: 2.8T total, 104B active parameters
➤ Input modalities: Text and image
➤ Weights and license: Weights currently not published, Moonshot is expected to release the weights soon after launch
➤ Pricing: $2/$10/$0.20 per 1M input/output/cache hit tokens, 33% lower than Kimi K3 on all three rates.
Congratulations to the team at @Kimi_Moonshot on the release!
اللي مش شايف ان هيبقى فيه wave عكس الاتجاه السائد للكلاود يبقى اعمى
الAI هيخلي الناس تشد في شعرها و بيزنس كتير هتتشمع ع حسه from security perspective
اي شركة محترمة هتحاول تhost it locally
اسعار الرامات اتضربت في خمسين اخر سنتين بسبب الAI و تعاقدات الداتا سنترز اللي استحوذت
🚨BREAKING: RAILWAY has just disclosed MAJOR security incident.
Customer API keys, database passwords, and other sensitive credentials may have been EXPOSED TO HACKERS.
Railway is urging ALL customers to rotate their credentials IMMEDIATELY.
And in an extraordinary move, the cloud hosting giant has PAUSED ALL DEPLOYMENTS while it investigates.
Here's what we know:
→ POTENTIAL SECURITY BREACH: Railway detected unauthorized activity involving systems storing customer environment variables.
→ DECRYPTED SECRETS EXPOSED: Attackers may have accessed encrypted environment variables AND obtained the decrypted contents of some of them.
→ CRITICAL CREDENTIALS AT RISK: Database passwords, cloud credentials, API keys, and private registry credentials could all be compromised.
→ ALL DEPLOYMENTS HALTED: Railway has suspended builds and deployments across its platform. Its dashboard and API are also experiencing major disruptions.
→ EMERGENCY WARNING TO EVERY CUSTOMER: Railway is urging ALL users to immediately rotate stored credentials and investigate suspicious activity.
→ POSSIBLE COORDINATED ATTACK: Railway believes this could be part of a larger campaign targeting cloud platforms to compromise AI inference providers.
→ FULL SCOPE UNKNOWN: Railway says the impact may be limited to a subset of users, but it is still investigating.
Think about how serious this is.
The platform developers trust to PROTECT their application secrets may have exposed those very secrets to attackers.
And now they're being told to rotate credentials while new deployments are BLOCKED.
First PHP. Then Node.js. Then LiteLLM.
Now RAILWAY.
October 21, 2026 is turning into one of the most alarming days in cybersecurity this year.
If you use Railway, take action NOW.
ممكن برضو الخص الموضوع في كلمتين :
اي جهاز هتجيبه اقل من ١٢٨ جيجا هيشغلك موديلز اقل و context window صغير اوي
معاك open budget؟ m5 ultra with 256 gb
معاك اقل؟ mac m5 max chip with 128 gb
اي حاجة غير دول هندخل في تفاصيل كتير في مقارنة ما بين الreading and writing speed
بقالي فترة غرقان في بحر الlocal llms واختيار احسن بروداكت
بعد سماع هري وبيع كتير
احب اقول ان دي احسن تشانل تقدر تاخد منها معلومات تكنيكال كويسة لاي بروداكت طلع اخر سنة ونص للlocal llms
youtube.com/@kaiexplainsyt?s…
So basically Qwen 3.8 flash-next is so fast because he does “partitioning” with a lookup table with 6B parameters so he can quickly get the next generated token, hmmmm long live data engineering.
بقالي فترة غرقان في بحر الlocal llms واختيار احسن بروداكت
بعد سماع هري وبيع كتير
احب اقول ان دي احسن تشانل تقدر تاخد منها معلومات تكنيكال كويسة لاي بروداكت طلع اخر سنة ونص للlocal llms
youtube.com/@kaiexplainsyt?s…
