#mistralai — Public Fediverse posts

Mistral raises €3 billion Series D led by Samsung at over €21 billion valuation

Paris-based Mistral AI has closed a €3 billion Series D funding round at a post-money valuation of more…
#EuropeSays #Korea #KR #Samsung #Advent #ArthurMensch #ASML #BlackRock #Bpifrance #DSTGlobal #GeneralCatalyst #IndexVentures #MistralAI #NVIDIA #PSGEquity #SamsungElectronics #SamsungGroup #ScaleupEuropeFund #SeriesD
https://www.europesays.com/korea/151089/

https://www.europesays.com/cz/208698/ Rekordní suma na další expanzi. Evropský AI šampion uspěl u investorů #Business #Byznys #Firmy #Francie #investice #MistralAI #Počítač #UměláInteligence(AI)

Policy-aware moderation works best when the policy travels with the request.

I built Shieldstral into a Quarkus LangChain4j service as scored input and output guardrails. The tutorial covers the one-token classifier contract, vLLM with Podman, fail-closed outages, deterministic HTTP stubs, and real-model threshold calibration.

https://www.the-main-thread.com/p/shieldstral-quarkus-langchain4j-guardrails

#Quarkus #LangChain4j #AIGuardrails #MistralAI

Policy-aware moderation works best when the policy travels with the request.

I built Shieldstral into a Quarkus LangChain4j service as scored input and output guardrails. The tutorial covers the one-token classifier contract, vLLM with Podman, fail-closed outages, deterministic HTTP stubs, and real-model threshold calibration.

https://www.the-main-thread.com/p/shieldstral-quarkus-langchain4j-guardrails

#Quarkus #LangChain4j #AIGuardrails #MistralAI

Policy-aware moderation works best when the policy travels with the request.

I built Shieldstral into a Quarkus LangChain4j service as scored input and output guardrails. The tutorial covers the one-token classifier contract, vLLM with Podman, fail-closed outages, deterministic HTTP stubs, and real-model threshold calibration.

https://www.the-main-thread.com/p/shieldstral-quarkus-langchain4j-guardrails

#Quarkus #LangChain4j #AIGuardrails #MistralAI

Policy-aware moderation works best when the policy travels with the request.

I built Shieldstral into a Quarkus LangChain4j service as scored input and output guardrails. The tutorial covers the one-token classifier contract, vLLM with Podman, fail-closed outages, deterministic HTTP stubs, and real-model threshold calibration.

https://www.the-main-thread.com/p/shieldstral-quarkus-langchain4j-guardrails

#Quarkus #LangChain4j #AIGuardrails #MistralAI

Policy-aware moderation works best when the policy travels with the request.

I built Shieldstral into a Quarkus LangChain4j service as scored input and output guardrails. The tutorial covers the one-token classifier contract, vLLM with Podman, fail-closed outages, deterministic HTTP stubs, and real-model threshold calibration.

https://www.the-main-thread.com/p/shieldstral-quarkus-langchain4j-guardrails

#Quarkus #LangChain4j #AIGuardrails #MistralAI

📌 ChatGPT, Gemini e Copilot gestiscono i tuoi dati secondo regole che non controlli.
GDPRchat gira su server europei con Mistral AI: stessa interfaccia, giurisdizione europea. 👇
https://gomoot.com/gdprchat-il-nuovo-chatbot-ue-basato-su-mistral-ai-e-conforme-al-gdpr/

#FRITSAIAps #GDPRchat #Hetzner #MistralAI

@mistersql @MistralAI @BuyFromEU thanks for the reply hope to find some experts that know better the #mistralai way. Hope it's not only about #sovereignity

> Today, we are announcing #Amazon, #Meta, #Microsoft, #mistralai , and #Perplexity for the first time as they join our roster of partners, which includes #Google, #Ecosia, #Nomic, #Pleias, #ProRata, and #ReefMedia. All these organizations utilize #WikimediaEnterprise to integrate human-governed knowledge into their platforms at scale. By doing so, they help ensure that the work of our global volunteer community reaches billions of people with the accuracy and transparency that Wikipedia represents.

And that a good new for me.

#wikimedia #wikipedia #ai

https://enterprise.wikimedia.com/blog/wikipedia-25-enterprise-partners/

Mistral AI: Das französische KI-Startup im Vergleich zu OpenAI
Mistral AI ist einer der wenigen, wachsender Akteure auf dem europäischen KI-Markt und gilt als wichtigster Konkurrent zu OpenAI. Mit fortschrittlichen Modellen und der Chat-App
https://www.apfeltalk.de/magazin/news/mistral-ai-das-franzoesische-ki-startup-im-vergleich-zu-openai/
#KI #News #ArtificialIntelligence #Enterprise #Frankreich #Investment #KI #LeChat #Microsoft #MistralAI #Nvidia #OpenSource #OpenAI #Partnerschaften

Mistral AI seeks to raise funds at a $5 billion valuation, following recent investments from major firms like Microsoft and partnerships enhancing its AI models on Azure.

#MistralAI #Funding #Valuation #Microsoft #Partnerships #Azure #AI #Investment #TechNews #Innovation #AIModels #CloudComputing #Startup #TechFunding #AIInnovation #BusinessNews #FinancialNews #BigValuation #TechInvestment #FutureTech #AIProgress #MarketNews #CapitalRaise #StrategicPartnerships #GrowthPotential

Tried out the Mixtral-8x22B model that Mistral.ai has released a few days ago. Have to say that I'm impressed. It is the first open foundation model that can speak German flawlessly without grammar mistakes.

I haven't tested the fine-tunes that start to get released now, but I think it is clear that the foundation model is very strong.

In terms of hardware requirement, it will be out of reach for most people to run at home, I fear. Yet it is still in the same range as the biggest LLama2 based models. Being Mixture of Experts, it is comparable in speed to 70B Llama2 for me, and faster than the bigger Frankenstein models like Goliath (120B).

I would say, the minimum hardware requirement is 24 GB VRAM (e.g. Nvidia 3090/4090) and around 100 GB Memory. In such a setup, the CPU becomes the bottleneck anyway, and Mixture of Experts is a significant speedup in the phase when the output tokens are being generated.

But kudos to Mistral.ai for releasing such a great model!

#mistralai #mixtral8x22b #llm #ai

https://dataplatform.cloud.ibm.com/docs/content/wsj/getting-started/whats-new-wx.html?context=wx&locale=en

The mpt-7b-instruct2 model is deprecated.
Replaced by mixtral-8x7b-instruct-v01-q

Supported natural languages: English, French, German, Italian, Spanish

It only uses 13 billion active parameters for inferencing, which reduces costs and latency.

#MistralAI #LLMs #GenerativeAI #IBM #watsonx #mixtral #mixtral8xb #SDLC #SoftwareDevelopmentLifeCycle #RAG #RetrievalAugmentedGeneration