Enterprise Sales & Strategic Alliances | Sovereign Cloud, Data & AI Platforms | Europe @Exoscale 🔺
Zurich, Switzerland
Joined January 2011
- Tweets7.2K
- Following823
- Followers423
- Likes2.1K
A major bank is telling customers not to give banking logins to AI agents. Share them and you're responsible for what happens. An agent with your password is indistinguishable from you. Agents need their own scoped, revocable access.
techcrunch.com/2026/10/27/ba…
What agent access should look like: its own identity, read-only by default, limits per action, short-lived tokens you can revoke at any time, and a log of every call it makes.
Swiss AI rules are moving on two tracks: voluntary codes now, a consultation on binding rules for transparency, data protection, oversight and risk assessments planned for spring 2027.
Document where your AI runs, what data it uses and who can access it.
uvek.admin.ch/de/newnsb/s2TQ…
Anastasios Panagou retweeted
Announcing Artificial Analysis Intelligence Index v5, featuring 2 new benchmarks and 3 upgrades. For the first time, the majority of the Index is built on private datasets
The Artificial Analysis Intelligence Index is our composite measure of model capabilities across tasks reflecting real-world usage today. v5 is our biggest upgrade since v4 launched in January. With v5, 60% of the Index is now based on evaluations with private test sets or held-out answers.
Alongside Intelligence Index v5, we’re launching the Artificial Analysis Progress Index v1, which tracks model capability over time on one continuous scale.
New to the Intelligence Index:
➤ FrontierCode 2.0: Cognition's newly-launched private benchmark of real-world coding tasks, graded on whether a repository maintainer would merge the agent's change
➤ Terminal-Bench-Science: 70 agentic scientific research tasks, run in a terminal environment
Upgraded:
➤ HLE-Diamond (replacing Humanity's Last Exam): a refined 1,000-question version of Humanity's Last Exam, split evenly between reasoning and knowledge questions
➤ AA-Omniscience v1.1: An updated dataset with saturated and public set questions removed, new questions added, and ambiguous or erroneous questions revised
➤ CritPt v2: a private, 40-question hard subset of CritPt v2, revised by the CritPt authors and run with a code-execution tool
Staying in the Index: AA-Briefcase, AutomationBench-AA, GDP.pdf and Terminal-Bench 4.0
Removed from the Index: GDPval-AA v2.1, SciCode, and AA-LCR v1.1
Key results:
➤ Claude Opus 5.5 (max) leads at 58, ahead of GPT-6 Astra (max, 57) and GPT-6.1 Sol (max, 55): Anthropic and OpenAI hold the top five models. Gemini 4 Argon (high) is next at 50
➤ The open weights gap widens from 3 points to 12: Kimi K3.1 Pro (weights not yet released) is the top open weights model at 46. Among released weights, DeepSeek V4.1 Flash 1017 (max) and GLM-5.3 (max) lead at 32 - 26 points behind the leader and below Claude Haiku 5.5 (max, 36)
➤ Cost per task varies by more than 13x among the leading models: of the 10 highest-scoring models shown, GPT-6.1 Terra (max) is the cheapest at $0.96 per Intelligence Index task and Claude Fable 5.1 (max) the most expensive at $13.04
Anastasios Panagou retweeted
Small South European economies are undergoing the largest and fastest public debt reduction episodes in recent history based on sound fiscal policies and strong growth, not financial repression or inflation. One for the history books. 🇬🇷🇵🇹🇨🇾🇭🇷
Anastasios Panagou retweeted
Greece 🇬🇷 jumps to 10th!
Czechia 🇨🇿 falls to 12th!
Hungary 🇭🇺 climbs to 17th, Croatia 🇭🇷 to 19th!
Latvia 🇱🇻 climbs to 34th!
Anastasios Panagou retweeted
🏁 Taking Gemini 4 Argon for a spin in Google @Antigravity.
Learn how @genevieve__h transforms a desk photo into an interactive 3D racing game by using Argon to orchestrate subagents and leverage Nano Banana to generate assets for the build.
Anastasios Panagou retweeted
🇬🇷🔥 BOUZOUKI, GREEK PRIDE AND HELLAS!
The pre-game entertainment at the Australia Cup final brought a taste of Greece to Melbourne, with the bouzouki playing and Sfakinaki’s “Ego Eimai Ellinas” (“I Am Greek”) setting the mood before kick-off.
As South Melbourne FC, the pride of Greek Australians, prepared to face Melbourne Victory, the atmosphere was unmistakably Greek. 🇬🇷🎶
🏆 THE FINAL SCORE: SOUTH MELBOURNE 1–1 MELBOURNE VICTORY
Hellas won 4–3 on penalties! 🇬🇷⚽
From the bouzouki to the terraces, this was Hellenism on full display on one of Australian football’s biggest stages.
Εγώ είμαι Έλληνας! 🇬🇷
#SouthMelbourneFC #Hellas #EgoEimaiEllinas #GreekAustralians #GreekPride #AustraliaCup
Anastasios Panagou retweeted
Claude Managed Agents dynamic workflows are now available in public beta. It's a new type of multiagent orchestration, built for your most ambitious workloads.
A lead agent writes a plan that runs across many agents in phases, combining the results at the end.
Anastasios Panagou retweeted
Now in beta: composer predictions in Codex for Pro users.
Codex can now suggest your next message based on your conversation and how you talk to it.
One of the most loved new features we've ever tested internally.
Anastasios Panagou retweeted
Monthly API credits are now available on Max and Team plans: $100 on Max 5x, $200 on Max 20x, and up to $500 pooled on Team. Use them on any Claude model.
How it works and full terms: support.claude.com/en/articl…
We’re rolling out monthly Claude Platform API credits for Max and Team plans:
Max 5x: $100
Max 20x: $200
Team: up to $500, pooled
Works on any model, including Haiku 5.5, in your code or third-party harnesses.
How it works and full terms: support.claude.com/en/articl…
Anastasios Panagou retweeted
Frontier scale, built in Europe.
Nearly 4,000 NVIDIA Grace Blackwell Superchips powered the training of Mistral Large 4, now in public preview.
Congrats to the @MistralAI team 🙌
Meet Mistral Large 4, aka Le Chonk.
• 1T parameters, natively multimodal. 49B active.
It is the best open weights model from US or Europe on aggregated benchmarks.
• State-of-the-art on critical workloads, including cyber defense, manufacturing and finance and it surpasses closed frontier models on visual grounding.
• Forged in Europe end-to-end and is deployable from Europe via our own Mistral Cloud infrastructure.
• Available to all via API today. Working with cybersecurity partners privately.
Open weights release end of October.
Anastasios Panagou retweeted
Small bird, fast wings, Kolibri is here.
78B parameters. 3.46B active. Up to 1M tokens of context. Built in Europe.
Now the weights are yours. Run it on your own hardware, under Apache 2.0.
Anastasios Panagou retweeted
Random German lab appears from nowhere and drops SOTA open-weight model
Beats Qwen, Mistral, Nematron models with similar active parameter count
Trained from stratch
Small bird, fast wings, Kolibri is here.
78B parameters. 3.46B active. Up to 1M tokens of context. Built in Europe.
Now the weights are yours. Run it on your own hardware, under Apache 2.0.
Readers added context they thought people might want to know
x.com/kimmonismus/st…
It beats older generation models and already obsolete nemotron 3 super and mistral 4 small(not even 3.5 medium) and it beats Qwen 3.6 35B a3b(which is very small) not current gen Qwen 3.8 27B and flash next. And the benchmarks are selective, they do not demonstrate coding,etc.
A strong addition for Kubernetes on Exoscale SKS: 𝗔𝗶𝗿𝗹𝗼𝗰𝗸 𝗠𝗶𝗰𝗿𝗼𝗴𝗮𝘁𝗲𝘄𝗮𝘆 by Ergon brings API & application security directly into the cluster.
Now 𝗪𝗼𝗿𝗸𝘀 𝘄𝗶𝘁𝗵 𝗘𝘅𝗼𝘀𝗰𝗮𝗹𝗲 certified. Looking forward to our continued collaboration.
linkedin.com/posts/kubernete…
#Kubernetes #Exoscale
Sovereign AI gets practical when you control the agent, model and where inference runs.
@goose_oss + @Alibaba_Qwen on @Exoscale Dedicated Inference: open-source, model-agnostic, dedicated GPU, European zone, OpenAI-compatible endpoint.
exoscale.com/blog/sovereign-…
