Arena, the AI evaluation platform formerly known as LMArena, launched its Alignment Index on October 8, a benchmark built from real-world agent sessions that found 48% of code debugging sessions ended…
Arena raised a $200 million Series B at a $3.1 billion valuation on October 8th, co-led by Lightspeed Venture Partners and Khosla Ventures, and launched the Arena Alignment Index, a preview that compa…
Anthropic's Claude Opus 5.5 cut em dash usage by roughly 95%, from 15.2 per 1,000 words in Opus 5 to 0.8, according to an analysis by the AI benchmarking tool Arena of high-reasoning Text Arena respon…
Several X users reported that Google's Gemini 4 Pro is being tested on Arena under the model name 'Gemini 3.8 Flash', with one user citing a 256k output ceiling as evidence the model is Gemini 4 Pro. …
Researchers at UC Berkeley and Arena published HarnessTax on September 16, 2026, finding that swapping the harness around the same coding model changed task success by only about ±2 points on SWE-benc…
Arena's September 11th WebDev snapshot ranked OpenAI's GPT-6 Astra Max first with a score of 1800 from 2,281 votes, ahead of Anthropic's Claude Fable 5.1 Max at 1758 from 3,036 votes and Claude Opus 5…
OpenAI released its ChatGPT Images 2.5 image generation model on September 8, and the precision-focused Sunburst variant reached the top spot on Arena's text-to-image, single-image edit, and multi-ima…
Microsoft has launched MAI-Image-2.6, its latest first-party image-generation model, which now ranks No. 2 on Arena's text-to-image leaderboard and No. 3 for image editing. The model is available in M…
Microsoft AI's MAI-Image-2.5 model has appeared in different positions across published Arena leaderboard snapshots, with launch materials reporting No. 2 for image editing and No. 3 for text-to-image…
Microsoft AI's MAI-Image-2.6 model scored 1,336 points to rank second in Arena's text-to-image leaderboard on August 10th, 45 points behind OpenAI's GPT Image 2 Medium and 20 points ahead of Grok Imag…
A codenamed image generator called "mona-lisa-1" appeared on Arena's text-to-image leaderboard testing on August 9th, prompting speculation that OpenAI is evaluating a new GPT Image model, though its …
XAI released Imagine Image 2.0 on August 7th, adding targeted editing, multi-image references, and workflow templates to its Grok image generator, now generally available as Quality Mode on Grok Imagi…
Arena CEO Anastasios Angelopoulos said his AI-model evaluation company has interviewed job applicants who turned out to be AI-generated, not real people, and that the company is considering requiring …
DeepSeek V4 Flash, released by DeepSeek, is priced at 1/100th the cost of Fable and 1/60th of GPT-5.6 Sol while remaining competitive on benchmarks, according to Artificial Analysis and Arena leaderbo…
Arena launched AutoEval on July 30, giving new AI models a preliminary Arena Score within an hour by training reward models on millions of pairwise preferences, compared with days required for human v…
A group of 133 US tech companies led by Microsoft and Nvidia published an open letter urging policymakers to avoid blanket bans on open-weight AI models, expand compute access for researchers, and tak…
Arena has launched a factuality leaderboard ranking models on correctness of claims alongside human preference, built on more than 2 million labeled claims from real conversations. Arena told Lets Dat…
Anthropic launched Claude Opus 5, achieving an ECI of 159 and matching Fable 5 on SWE-ECI at 161, according to Epoch AI Research. Independent evaluations show Opus 5 outperforming Fable in coding and …
A developer tested Kimi K3, a new open AI model from Moonshot AI, for a week of real work. The model handled a 1M token context window by refactoring a 40,000-line codebase, and ranked #1 on Arena's F…
Beijing-based startup Moonshot released its Kimi K3 AI model, which tops Arena's ranking for front-end coding capability and costs half as much as OpenAI's GPT-5.6 Sol, according to Bank of America. T…