MAROKO133 Update ai: Mistral launches powerful Devstral 2 coding model including open sour

📌 MAROKO133 Eksklusif ai: Mistral launches powerful Devstral 2 coding model includ

French AI startup Mistral has weathered a rocky period of public questioning over the last year to emerge, now here in December 2025, with new, crowd-pleasing models for enterprise and indie developers.

Just days after releasing its powerful open source, general purpose Mistral 3 LLM family for edge devices and local hardware, the company returned today to debut Devstral 2.

The release includes a new pair of models optimized for software engineering tasks — again, with one small enough to run on a single laptop, offline and privately — alongside Mistral Vibe, a command-line interface (CLI) agent designed to allow developers to call the models up directly within their terminal environments.

The models are fast, lean, and open—at least in theory. But the real story lies not just in the benchmarks, but in how Mistral is packaging this capability: one model fully free, another conditionally so, and a terminal interface built to scale with either.

It’s an attempt not just to match proprietary systems like Claude and GPT-4 in performance, but to compete with them on developer experience—and to do so while holding onto the flag of open-source.

Both models are available now for free for a limited time via Mistral’s API and Hugging Face.

The full Devstral 2 model is supported out-of-the-box in the community inference provider vLLM and on the open source agentic coding platform Kilo Code.

A Coding Model Meant to Drive

At the top of the announcement is Devstral 2, a 123-billion parameter dense transformer with a 256K-token context window, engineered specifically for agentic software development.

Mistral says the model achieves 72.2% on SWE-bench Verified, a benchmark designed to evaluate long-context software engineering tasks in real-world repositories.

The smaller sibling, Devstral Small 2, weighs in at 24B parameters, with the same long context window and a performance of 68.0% on SWE-bench.

On paper, that makes it the strongest open-weight model of its size, even outscoring many 70B-class competitors.

But the performance story isn’t just about raw percentages. Mistral is betting that efficient intelligence beats scale, and has made much of the fact that Devstral 2 is:

• 5× smaller than DeepSeek V3.2

• 8× smaller than Kimi K2

• Yet still matches or surpasses them on key software reasoning benchmarks.

Human evaluations back this up. In side-by-side comparisons:

• Devstral 2 beat DeepSeek V3.2 in 42.8% of tasks, losing only 28.6%.

• Against Claude Sonnet 4.5, it lost more often (53.1%)—a reminder that while the gap is narrowing, closed models still lead in overall preference.

Still, for an open-weight model, these results place Devstral 2 at the frontier of what’s currently available to run and modify independently.

Vibe CLI: A Terminal-Native Agent

Alongside the models, Mistral released Vibe CLI, a command-line assistant that integrates directly with Devstral models. It’s not an IDE plugin or a ChatGPT-style code explainer. It’s a native interface designed for project-wide code understanding and orchestration, built to live inside the developer’s actual workflow.

Vibe brings a surprising degree of intelligence to the terminal:

• It reads your file tree and Git status to understand project scope.

• It lets you reference files with @, run shell commands with !, and toggle behavior with slash commands.

• It orchestrates changes across multiple files, tracks dependencies, retries failed executions, and can even refactor at architectural scale.

Unlike most developer agents, which simulate a REPL from within a chat UI, Vibe starts with the shell and pulls intelligence in from there. It’s programmable, scriptable, and themeable. And it’s released under the Apache 2.0 license, meaning it’s truly free to use—in commercial settings, internal tools, or open-source extensions.

Licensing Structure: Open-ish — With Revenue Limitations

At first glance, Mistral’s licensing approach appears straightforward: the models are open-weight and publicly available. But a closer look reveals a line drawn through the middle of the release, with different rules for different users.

Devstral Small 2, the 24-billion parameter variant, is covered under a standard, enterprise- and developer-friendly Apache 2.0 license.

That’s a gold standard in open-source: no revenue restrictions, no fine print, no need to check with legal. Enterprises can use it in production, embed it into products, and redistribute fine-tuned versions without asking for permission.

Devstral 2, the flagship 123B model, is released under what Mistral calls a “modified MIT license.” That phrase sounds innocuous, but the modification introduces a critical limitation: any company making more than $20 million in monthly revenue cannot use the model at all—not even internally—without securing a separate commercial license from Mistral.

“You are not authorized to exercise any rights under this license if the global consolidated monthly revenue of your company […] exceeds $20 million,” the license reads.

The clause applies not only to the base model, but to derivatives, fine-tuned versions, and redistributed variants, regardless of who hosts them. In effect, it means that while the weights are “open,” their use is gated for large enterprises—unless they’re willing to engage with Mistral’s sales team or use the hosted API at metered pricing.

To draw an analogy: Apache 2.0 is like a public library—you walk in, borrow the book, and use it however you need. Mistral’s modified MIT license is more like a corporate co-working space that’s free for freelancers but charges rent once your company hits a certain size.

Weighing Devstral Small 2 for Enterprise Use

This division raises an obvious question for larger companies: can Devstral Small 2 with its more permissive and unrestricted Apache 2.0 licensing serve as a viable alternative for medium-to-large enterprises?

The answer depends on context. Devstral Small 2 scores 68.0% on SWE-bench, significantly ahead of many larger open models, and remains deployable on single-GPU or CPU-only setups. For teams focused on:

• internal tooling,

• on-prem deployment,

• low-latency edge inference,

  …it offers a rare combination of legality, performance, and convenience.

But the performance gap from Devstral 2 is real. For multi-agent setups, deep monorepo refactoring, or long-context code analysis, that 4-point benchmark delta may understate the actual experience difference.

For most enterprises, Devstral Small 2 will serve either as a low-friction way to prototype—or as a pragmatic bridge until licensing for Devstral 2 becomes feasible. It is not a drop-in replacement for the flagship, but it may be “good enough” in specific production slices, particularly when paired with Vibe CLI.

But because Devstral Small 2 can be run entirely offline — including on a single GPU machine or a sufficiently specced laptop — it unlocks a …

Konten dipersingkat otomatis.

🔗 Sumber: venturebeat.com

📌 MAROKO133 Eksklusif ai: US engineers design AI bionic hand that grips with human

Engineers at the University of Utah have given a bionic hand a mind of its own.

By equipping a commercial prosthetic with pressure and proximity sensors and training an AI neural network on natural grasping movements, the team created a hand that grips more intuitively and securely.

Study participants were able to perform everyday tasks—such as picking up small items or raising a cup—with greater precision and less mental effort, without extensive practice.

According to researchers, the breakthrough points to a future where prosthetics feel and function more like natural limbs.

In May 2025, Korean researchers presented an ultra-light robotic hand with shape-adaptive grips, precise fingertip control, and thumb flexibility, powered by a single actuator.

Dexterity meets AI

Everyday tasks like reaching for a mug, picking up a pencil, or shaking someone’s hand rely on the brain’s ability to control finger movements instinctively. For people using prosthetic arms and hands, this natural dexterity is often lost. Even with advanced robotic prostheses, performing simple actions requires extra mental effort, as users must consciously control each finger to grasp objects.

According to the team, a major challenge is that most commercial bionic hands lack the sense of touch that allows humans to grip intuitively. Yet dexterity involves more than sensory feedback—our brains also subconsciously model and predict hand-object interactions, enabling reflexive, precise movements.

“As lifelike as bionic arms are becoming, controlling them is still not easy or intuitive. Nearly half of all users will abandon their prosthesis, often citing their poor controls and cognitive burden,” said Marshall Trout, a postdoctoral researcher in the Utah NeuroRobotics Lab, in a statement.

To tackle these challenges, researchers at the University of Utah partnered with TASKA Prosthetics to enhance a commercial robotic hand. They equipped the fingers with custom fingertips that detect pressure and include optical proximity sensors, mimicking the subtle sense of touch. The sensors are sensitive enough to detect something as light as a cotton ball landing on the hand.

The team then trained an artificial neural network on the proximity data, teaching the hand to automatically adjust each finger’s position for a stable, precise grip. With each finger operating independently yet in coordination, the hand can form an optimal grasp on virtually any object.

Researchers claim this combination of touch replication and AI-driven movement allows the prosthetic to function more naturally, reducing mental strain and improving everyday usability.

Intuitive hand control

As the development progressed, one challenge remained: ensuring the prosthetic could adapt if the user didn’t intend to grasp an object in the AI-predicted manner, such as when they wanted to release it.

To solve this, the researchers developed a bioinspired system that shares control between the user and the AI, carefully balancing human intent with machine precision. The AI augments natural movements, enhancing grip accuracy while reducing the mental effort required to complete tasks.

The team tested the system with four participants who had amputations between the wrist and elbow. In addition to performing better on standardized assessments, participants successfully completed everyday tasks that require fine motor control. Activities as simple as drinking from a plastic cup, which demand precise pressure to avoid dropping or crushing it, became manageable.

According to researchers, combining AI assistance with human intent enabled the prosthetic hand to offer a more intuitive, natural experience, allowing users to perform daily tasks with less cognitive strain and greater confidence.

“By adding some artificial intelligence, we were able to offload this aspect of grasping to the prosthesis itself. The end result is more intuitive and more dexterous control, which allows simple tasks to be simple again,” said Jacob A. George, a postdoctoral researcher in the Utah NeuroRobotics Lab, in a statement.

The study team is exploring implanted neural interfaces that would allow users to control prostheses with their minds while restoring a sense of touch.

Their next steps involve integrating these technologies so that the enhanced sensors improve tactile function and the intelligent prosthetic can operate seamlessly with thought-based control.

🔗 Sumber: interestingengineering.com