#mistral · Cascadity

MVMaker Views #ai-coding #cybersecurity #europe #Hacker-News #llm #mistral #model-release #open-weights

Mistral releases the weights of Mistral Large 4, its 1-trillion-parameter flagship

Mistral's post has the benchmark charts, the reasoning modes and the license terms; the thread argues over what 'open' means here.

Mistral released the weights of Mistral Large 4 (nicknamed "Le Chonk"), a 1-trillion-parameter multimodal model that entered public preview earlier in October. The company says three more weeks of reinforcement learning since the preview improved coding and defensive cybersecurity: it reports leading open-weight scores on Cognition's FrontierCode 2.0 (19.7%) and Terminal-Bench 4.0 (41%), and 51% on the Artificial Analysis Cyber Index, with a 96.6% refusal rate on malicious cyber prompts. These are vendor-reported results. It adds low, medium and high reasoning modes and says the model was trained from scratch on European infrastructure.

Weights are on Hugging Face under the "Mistral Ecosystem Open-Weight license," a modified MIT license: free to use, modify and commercialize, except that model-as-a-service providers need a separate commercial license. API pricing is $0.70 per million input tokens and $2.10 per million output tokens.

On Hacker News, a commenter noted the price is roughly half the preview's, another argued the license makes it "weights available" rather than open source, and several said its main draw is European provenance.

Read it at the source

Mistral AIMistral Large 4 (Le Chonk) is now open weight | Mistral ↗ Hacker NewsDiscussion: We're releasing the weights of Mistral Large 4 Independent ↗

Cascadic Analysis 2

UndertowBenchmark scores can overstate real-world reliability

What hidden risk could pull against this, even if the news is good?

Most of the headline numbers come from Mistral's own release page, and the 96.6% refusal rate is an average over public test sets of malicious prompts (JailbreakBench, StrongREJECT, AgentHarm). Known prompts are what a determined user is least likely to send. Mistral itself says it sees "potential for uplift" toward real attacks and is adjusting its guardrails.

UndertowAutonomous cyber capability scales attackers enormously

What hidden risk could pull against this, even if the news is good?

Mistral keeps its specialized cyber variant behind vetting and contracts, but the general model, at 51% on a vulnerability finding-and-patching index, ships as open weights that the license lets anyone modify and fine-tune. Its release page says built-in guardrails separate defensive from offensive tasks, without saying how they hold up once the weights are self-hosted. A downloaded model can't be recalled if that line proves thin.

Rabbit Holes 2

Open-source artificial intelligence

After exploring Mistral Large 4 (Le Chonk) is now open weight | Mistral, for the debate over what counts as 'open', read Open-source artificial intelligence.

Covers the Open Source Initiative's definition of open-source AI (free to use, study, modify and share, including data, code and parameters) and how it differs from releasing weights alone.

Wikipedia · Wade · 5 min

0
Open weights

After exploring Mistral Large 4 (Le Chonk) is now open weight | Mistral, for what 'open weights' does and doesn't mean, read Open weights.

Explains open weights: publicly released learned parameters of an AI model, which let others download and run it, with modification and redistribution depending on the license.

Wikipedia · Wade · 5 min

0

MVMaker Views #cern #europe #Hacker-News #mistral #speech-ai #speech-recognition #text-to-speech #time-series #voice-agents

Mistral launches Voxtral 3 speech models and a voice-agent SDK covering 50+ languages

Mistral's post has benchmark charts against Whisper and Gemini, plus pricing for each model.

Mistral announced Voxtral 3, a family of speech-to-text and text-to-speech models plus a voice-agent SDK that combines listening, reasoning, tool calling and speaking in one real-time pipeline. Mistral says Voxtral Small Transcribe 3 handles 50 languages, identifies who spoke when and outperforms Whisper, Scribe v2 and Gemini 3.5 Transcribe; the 4B-parameter Mini version costs $0.0035 per minute. Voxtral Mini TTS 3 adjusts emotion and pace across 35+ languages, and a Pro Voice feature adapts a speaker's voice from 30 minutes of audio. It is offered through Mistral's EU-hosted API or for self-deployment.

On Hacker News, commenters joked that the demo's quick French dermatology booking was the least realistic part, regretted that the models aren't open weights, and worried that phone agents wired into company systems could be talked into things; others traded notes on local transcription tools.

Read it at the source

Mistral AIVoxtral 3: Voice Agent SDK & Speech Models in 50+ Languages Independent ↗ Hacker NewsDiscussion: Voxtral 3 Independent ↗ Mistral AITimestral: Mistral's time-series foundation models Independent ↗ CERNCERN and Mistral join forces to accelerate AI for science Independent ↗

Cascadic Analysis 2

UndertowConfused-deputy / excessive-agency problem

What hidden risk could pull against this, even if the news is good?

Mistral says the voice-agent SDK lets agents retrieve information, interact with business systems and complete tasks during live phone or web calls. Hacker News commenters worried such agents could be talked into things: the caller is an untrusted stranger, but the agent acts with the company's access, so what it is allowed to do on a call matters more than how natural it sounds.

UndertowBenchmark scores can overstate real-world reliability

What hidden risk could pull against this, even if the news is good?

The claim that Voxtral Small Transcribe 3 beats Whisper, Scribe v2 and Gemini 3.5 Transcribe is Mistral's comparison. Speech recognition is especially sensitive to accents, background noise and phone-line audio, which a vendor's test set may not represent; Hacker News commenters joked that the demo's quick French dermatology booking was its least realistic part.

Rabbit Holes 2

Speaker diarisation

After exploring Voxtral 3: Voice Agent SDK & Speech Models in 50+ Languages, for the "who said what" feature, read about Speaker diarisation.

The process of splitting an audio stream into segments by speaker, answering the question "who spoke when?"

Wikipedia · Wade · 5 min

0
Large Hadron Collider

After exploring CERN and Mistral join forces to accelerate AI for science, the machine whose collisions the model will learn from is the Large Hadron Collider.

Describes the world's largest and highest-energy particle accelerator, built by CERN between 1998 and 2008.

Wikipedia · Swim · 30 min

0