Microsoft AI says image Elo rose from 864 to 1147 in nine months

MMicrosoft AI9/23/2026, 10:05:28

Microsoft AI says no lab has reached the image frontier faster. It reports an Elo rise from 864 to 1147 in nine months, which it says equals 377 points a year and 1.6 times the next-fastest climb on the board.

#microsoft-ai #image-generation #elo #benchmarks

View original post ↗

Previous articleMAI-Image-2.6 can be tried on Foundry and OpenRouter Next articleMicrosoft AI says MAI is hill-climbing image generation 61% faster than any other lab

Related AI news

Microsoft AI says MAI is hill-climbing image generation 61% faster than any other lab

Microsoft AI says MAI gained 283 Elo on text-to-image in 10 months, putting its image-generation hill-climbing pace 61% ahead of any other lab while optimizing the Pareto frontier.

MMicrosoft AI•9/23/2026, 10:02:48•

#microsoft-ai #mai #image-generation #text-to-image

Microsoft AI brings MAI-Code-1.1-Flash to GitHub Copilot for local execution

Developers will be able to run agentic coding tasks on-device using MAI-Code-1.1-Flash inside GitHub Copilot without paying inference fees for local model calls.

MMicrosoft AI•10/7/2026, 13:35:03•

#microsoft-ai #github-copilot #on-device-ai #mai-code

MAI-Image-2.6 can be tried on Foundry and OpenRouter

Microsoft AI says MAI-Image-2.6 is available to try on Foundry and OpenRouter.

MMicrosoft AI•9/23/2026, 10:00:39•

#mai-image-2.6 #microsoft-ai #foundry #openrouter

DeepLearning.AI details DeepSeek cache reduction and Flash benchmark results in The Batch

In an issue of The Batch, DeepLearning.AI analyzed DeepSeek's architecture, reporting an 890-byte cache per token that is 437 times smaller than DeepSeek-V1. The analysis also notes a 25% compute increase per output token from 4K to 1M input tokens and highlights Flash outperforming V4-Pro on the AA Index.

DDeepLearning.AI•10/7/2026, 17:02:37•

#deeplearning.ai #deepseek #benchmarks #llm-inference

Arena.ai evaluates Jev Router across more than 4,700 agentic sessions on Agent Arena

Arena.ai evaluated typesafeai's Jev Router across more than 4,700 agentic sessions on Agent Arena, finding that while it does not improve on the current Pareto frontier, it demonstrates strong steerability in response to user feedback.

AArena.ai•10/7/2026, 15:33:08•

#arena.ai #jev-router #typesafeai #benchmarks

Reset history AI News FAQ Usage limits RSS feeds API MCP Skill

News is compiled from public information. The data is for reference only, and this site has no affiliation with OpenAI.

Data & original project: Did Codex Reset · Licoy