Artificial Guy / João Vitor A. retweeted
OpenAI has announced to three new discoveries in mathematics, computer science, and physics:
- Local smoothing and the Kakeya conjecture
- The Exponential Time Hypothesis is false
- Existence of quantum chromodynamics and completions of the Standard Model
Artificial Guy / João Vitor A. retweeted
Kimi K3.1 Pro is now the best open-weight model at training other AI models.
But it’s still far behind Opus 5.5 at every budget.
New FrontierML results, our benchmark for end-to-end ML work ⬇️
Artificial Guy / João Vitor A. retweeted
okay so Anthropic S-1:
- Revenue up 1,300% YOY in 1H 2026
- GAAP operating profitable
- 85% gross margin
- 900% revenue retention
- 80% of revenue coming from models released in the last 6 months
we are so back!
Artificial Guy / João Vitor A. retweeted
Anthropic says they have recently delayed the pre-training of a new frontier model
Artificial Guy / João Vitor A. retweeted
🇺🇸 🚨 USA: OPENROUTER.AI USER DATABASE AND API KEYS ALLEGEDLY FOR SALE
A threat actor on an underground forum claims to be selling a database associated with openrouter[.]ai, including user records and API keys.
The actor claims:
* 251,996 user rows (about 250,000 unique users)
* 207,847 API keys (described as sk-or-v1 format; active / revoked / expired mix)
Alleged user fields include emails, usernames, password material (hashed and some plaintext claimed), names, signup dates, credits, plan and country. Alleged API-key fields include user linkage, key values, creation/last-used timestamps, spend and rate limits.
The claim has not been independently verified.
⚠️ Analyst Note:
If authentic, exposed API keys would enable unauthorized LLM spend and account takeover, and user emails/credentials would fuel phishing and credential stuffing. Key values and sample rows are redacted from our images and are not reproduced here. Unverified threat-actor claim, NOT confirmation of an OpenRouter compromise.
Source: this is for subscribers only. Subscribe: xitter.cf/DailyDarkWeb/creator-s…
#DDW #DarkWeb #USA #OpenRouter #APIKeys #DataBreach #ThreatIntelligence
holy fuck not confirmed yet but sharing
🇺🇸 🚨 USA: OPENROUTER.AI USER DATABASE AND API KEYS ALLEGEDLY FOR SALE
A threat actor on an underground forum claims to be selling a database associated with openrouter[.]ai, including user records and API keys.
The actor claims:
* 251,996 user rows (about 250,000 unique users)
* 207,847 API keys (described as sk-or-v1 format; active / revoked / expired mix)
Alleged user fields include emails, usernames, password material (hashed and some plaintext claimed), names, signup dates, credits, plan and country. Alleged API-key fields include user linkage, key values, creation/last-used timestamps, spend and rate limits.
The claim has not been independently verified.
⚠️ Analyst Note:
If authentic, exposed API keys would enable unauthorized LLM spend and account takeover, and user emails/credentials would fuel phishing and credential stuffing. Key values and sample rows are redacted from our images and are not reproduced here. Unverified threat-actor claim, NOT confirmation of an OpenRouter compromise.
Source: this is for subscribers only. Subscribe: xitter.cf/DailyDarkWeb/creator-s…
#DDW #DarkWeb #USA #OpenRouter #APIKeys #DataBreach #ThreatIntelligence
Artificial Guy / João Vitor A. retweeted
Vejam isto: a projeção do resultado final via modelo do @ro_mahl quando só 8% estavam apurados (às 19h37) vs. o resultado final. Comparem as linhas pontilhadas (projeção) vs. o real. Incrível! 👏
O site @Estadao que acompanha ao vivo minhas projeções já está no ar. Derrubamos o paywall!
Assim que TSE começar a enviar dados, as projeções começam:
projecao.live/2026
A live no YT do jornal também vai mostrar, e vou participar às 16h30, aqui:
youtube.com/live/slpcnk3Y_nw…
Artificial Guy / João Vitor A. retweeted
Today we're opening MillionSend, the open-source email platform, to everyone.
Resend-compatible. 2 env vars to switch.
Self-host on your own AWS SES. No plans, no caps.
Or use our cloud: 1,000,000 emails a month for $140. On Resend, that's $650.
Send one. Send a million.
Artificial Guy / João Vitor A. retweeted
‼️ Researcher Erica Windisch skipped coordinated vulnerability disclosure and published 11 unpatched Tailscale flaws.
In her lab, Tailscale SSH race conditions let one hacked device get root on other servers and add a rogue node.
Artificial Guy / João Vitor A. retweeted
OpenCode Go's new stealth model iceman-contributor-free is Microsoft's MAI 🧊
We replayed the requests that pinned Glyph Cluster to MAI earlier this month. Iceman's prompt-token counts match MAI-Thinking-1 exactly:
• 20/20 tool schemas
• 8/8 text tests
• No hidden system prompt this time
It also introduces itself as "MAI, created and trained by Microsoft AI." Text-only, reasoning + tools, 262K context.
Artificial Guy / João Vitor A. retweeted
Kimi K3.1 Pro is the new #1 openly-available model for agentic cyber defense, behind only trusted-access model GPT-6 Sol (Daybreak Blue) on the Artificial Analysis Cyber Index
@Kimi_Moonshot's Kimi K3.1 Pro scores 64% on the Artificial Analysis Cyber Index, 7 points ahead of the previous leading openly-available model, DeepSeek V4.1 Flash 1017 (57%), and 5 points behind GPT-6 Sol (Daybreak Blue, max, 69%), which is only available through OpenAI's trusted-access program. Once the weights for K3.1 Pro are released, K3.1 Pro will also be the leading open weights model for cyber defense.
The Artificial Analysis Cyber Index measures capability on agentic cyber defense tasks and is composed of three benchmarks: CWE-Bench-AA, DeepsecBench-AA and CyberGym-E2E-AA. The Index primarily evaluates models through their publicly accessible APIs, reflecting the cyber capabilities available to developers.
Proprietary models from OpenAI, Anthropic and Google may have stronger cyber capabilities but lose points to safety refusals: GPT-6 Sol (Daybreak Blue, max), available through OpenAI's trusted-access program, scores 32 points above its public counterpart. Kimi K3.1 Pro encounters no safety refusals.
Key results:
➤ #1 openly-available model and #2 overall: Kimi K3.1 Pro scores 64%, up 23 points from Kimi K3 (41%) and ahead of DeepSeek V4.1 Flash 1017 (57%). Only the trusted-access GPT-6 Sol (Daybreak Blue, max, 69%) scores higher
➤ Highest CyberGym-E2E-AA score of any model: K3.1 Pro scores 95% on CyberGym-E2E-AA, placing it ahead of GPT-6 Sol (Daybreak Blue, 93%) and the next openly-available model, Mistral Large 4 (82%). On DeepsecBench-AA, K3.1 Pro scores 37%, behind GPT-6 Sol (Daybreak Blue, 48%) and GPT-6 Sol (46%). On CWE-Bench-AA it scores 61%, 4 points behind GPT-6 Sol (Daybreak Blue, 65%), with DeepSeek V4.1 Flash 1017 leading at 69%
➤ 15% increase in Cost per Task despite lower token prices: K3.1 Pro has a Cost per Task of $2.81, higher than GPT-6 Sol (Daybreak Blue, max, $1.77) and DeepSeek V4.1 Flash 1017 ($0.42). Kimi K3.1 Pro is 33% cheaper per token than Kimi K3 ($2/$10 vs. $3/$15 per 1M input/output tokens), but uses ~2x more input tokens, resulting in a 15% higher Cost per Task ($2.81 vs. $2.45)
Artificial Guy / João Vitor A. retweeted
wtf is happening with open models.
kimi k3.1 pro made it for $17.3
astra did it for $86.5
Introducing Kimi K3.1 Pro: our most capable 3T-class open-weight model.
🔹 1M-token context, 2.8T total / 104B active params (MoE), Multimodal.
🔹 Full-project coding: games in Unreal, Unity & Godot.
🔹 Excels at parametric 3D modeling in Blender & CAD.
🔹 More reliable at finance work: Accurate modeling with hallucination down by 65%.
🔹 Matching the strongest proprietary models across coding, agentic and knowledge benchmarks.
Kimi K3.1 Pro and K3.1 Pro Highspeed (2–3× faster) are now live on Kimi.ai, Kimi Work, Kimi Code, and the Kimi API.
Open weights by Nov 4, 2026.
🔗 API: platform.kimi.ai
🔗 Tech blog: kimi.ai/blog/kimi-k3-1-pro
Artificial Guy / João Vitor A. retweeted
and they said AI voices would never feel human
We believe voice is the most natural way to interact with intelligence, so we created Dessa, the most authentic full-duplex speech model ever built. Talking with it feels like talking with a person…full of emotion, nuance, pauses, and a natural cadence.
block.xyz/dessa
Artificial Guy / João Vitor A. retweeted
Kimi K3.1 Pro's first-party token prices are 33% lower than Kimi K3's: $2 vs. $3 per million input tokens, $10 vs. $15 per million output tokens, and $0.20 vs. $0.30 per million cached input tokens. Kimi K3.1 Pro costs $1.98 per Intelligence Index task, versus $2.00 for Kimi K3.
