AI models News

Read the latest aI models news, analysis, and industry updates on AI Wire Media.

ElevenLabs Scribe v2 Tops Artificial Analysis AA-WER 2.0 Speech-to-Text Benchmark

Artificial Analysis has released version 2.0 of its AA-WER speech-to-text benchmark, which measures the accuracy of speech recognition models. In the overall ranking, ElevenLabs’ Scribe v2 takes first place with a word error rate of just ...
Perplexity Releases Open-Source Embedding Models That Rival Google and Alibaba

AI search engine Perplexity has introduced two new text-embedding models that aim to match or outperform Google’s and Alibaba’s offerings while using only a fraction of the usual memory footprint. Both models are open source.
Google Launches Gemini 3.1 Pro With Major Reasoning Gains

With Gemini 3.1 Pro, Google aims to significantly strengthen the core intelligence of its model family. On a demanding reasoning benchmark, performance has more than doubled compared with its predecessor. That said, benchmarks are still just benchmarks.
Google Upgrades Gemini 3 Deep Think, Outperforms GPT-5.2 and Claude

Google has upgraded its Gemini 3 Deep Think reasoning mode, positioning the tool as a solution for complex problems in science and engineering.
Gemini 3 Deep Think Upgrade Sets New Reasoning Benchmarks

Google DeepMind has upgraded its specialized reasoning mode, Gemini 3 Deep Think, and is making it available through the Gemini app as well as via an API in an early-access program on Vertex AI. The upgrade is designed to support complex tasks in science, research, and
...
OpenAI Retires GPT-4o and Other Legacy Models in ChatGPT

OpenAI will discontinue several older AI models in ChatGPT on February 13, 2026, including GPT-4o, GPT-4.1, GPT-4.1 mini, and o4-mini. The models will remain available in the API for the time being. The company officially cites declining usage as the reason for the move: only 0.1% of users
...
Zhipu AI Unveils GLM-5, an Open-Source Model Challenging GPT-5.2 and Claude Opus 4.5

With GLM-5, Chinese AI company Zhipu AI has introduced an open-source model designed to compete with Claude Opus 4.5 and GPT-5.2 in coding and agent-based tasks.
ByteDance Launches Seedance 2.0, a Multimodal AI Video Generation Model

ByteDance has released a new version of its AI video generation model, Seedance 2.0. The model supports eight languages.
Claude Opus 4.6 Tops Artificial Analysis Intelligence Index

Claude Opus 4.6 is currently the best AI model—at least until Artificial Analysis benchmarks OpenAI’s Codex 5.3, which is expected to take the lead in coding tasks. For now, Anthropic’s new model ranks first in the Artificial Analysis Intelligence Index, which
...
Anthropic Launches Claude Fast Mode with 6× Higher Pricing

Anthropic has introduced a new Fast Mode for Claude, offering significantly lower latency at a substantially higher cost. For Claude Opus 4.6, Fast Mode is priced at up to six times the standard rate.
Waymo Unveils Genie-3–Based World Model for Autonomous Driving

Waymo has unveiled a generative simulation model for autonomous driving based on Google DeepMind’s Genie 3. It is designed to generate hyper-realistic scenarios that would be
...
Anthropic Releases Claude Opus 4.6 With 1M-Token Context Window and Benchmark Gains

Anthropic has introduced Claude Opus 4.6, a new flagship model. For the first time, it features a one-million-token context window, and is designed
...
OpenAI Launches GPT-5.3-Codex, Setting New Benchmarks in AI Coding Performance

OpenAI has released its latest coding model, GPT-5.3-Codex. According to the company, it combines the coding capabilities of GPT-5.2-Codex with the reasoning and knowledge strengths of GPT-5.2, while being 25% faster than
...
Mistral Launches Voxtral Transcribe 2 to Undercut Speech-to-Text Rivals on Price

Mistral AI aims to undercut competitors on price in speech recognition with Voxtral Transcribe 2. The second generation of its speech-to-text models starts at $0.003 per minute and, according to Mistral, delivers higher accuracy than models such
...
Meta Completes Pretraining of “Avocado,” Its Most Powerful AI Foundation Model Yet

Meta has completed the pretraining of its new AI model, codenamed “Avocado,” according to an internal memo.
Kuaishou Releases Kling AI 3.0 With Advanced Multimodal Video Generation

Chinese developer Kuaishou has unveiled the third version of its video generation model, Kling AI.
Alibaba Launches Qwen3-Coder-Next, a Compact Open-Weight Model for Coding Agents

Alibaba has unveiled Qwen3-Coder-Next, a new open-weight AI model designed for coding agents and local development. The model was trained on 800,000 verifiable tasks executed in runnable environments. Despite its relatively
...
Google’s Gemini Models Lead AI Rankings in Social and Strategy Games

Google’s Gemini models are currently outperforming rivals in social and strategic games. Google DeepMind, together with Kaggle, has expanded its Game Arena benchmark with two new titles: Werewolf and ...
Perplexity Signs $750M Azure Deal With Microsoft

AI search startup Perplexity has signed a $750 million contract with Microsoft to use its Azure cloud services, Bloomberg reports, citing people familiar with the matter. The ...
Google DeepMind Adds Agentic Vision to Gemini 3 Flash for Active Image Analysis

Google DeepMind equips its Gemini 3 Flash model with a new capability called “Agentic Vision.” The model is designed to actively investigate images rather than merely observe them passively — although the feature does not yet work fully automatically in all cases.

Página 1 de 2