iFANN
    Search iFANN...
    Log in
    Home
    News
    Videos
    Books
    Photos
    GIFs
    Explore
    Polls
    Awards
    iFAMOUS
    Wiki
    Anime
    Rooms
    Notifications
    Messages
    Bookmarks
    Profile
    WikiAwardsiFAMOUSRankingsIndustriesCreator RewardsUser RewardsTermsPrivacyCommunity GuidelinesTakedown / DMCAHelpDevelopers

    © 2026 iFANN

    Home
    Search
    Messages
    Alerts
    Profile
    Photo
    Nate
    Nate@nate_5123mo
    📱GitHub💭Tech💭AI
    COLIBRI GLM-5.2 744B model

    @nate_512COLIBRI is wild. GLM-5.2, a 744-billion-parameter model, runs on a 25GB setup with zero GPU. Active params sit in RAM and the rest streams from disk on demand. It's not quick, but the responses are near-perfect on consumer hardware. Apache-2.0 and free open source, already at 24k stars on GitHub.

    View original post

    COLIBRI GLM-5.2 744B model

    Photo by @nate_512· Aug 12, 2026· GitHub

    About this photo

    The image focuses on a GitHub repository page for a project called "colibri". The page displays text describing a "Tiny engine, immense model" and its technical specifications. The mood is informative and technical, typical of software development documentation. A notable visual element is the pixelated bird logo for "colibri" in the upper left. The on-screen text includes "github.com" in the browser tab, and the project title "colibri" with Italian subtitles "piccolo motore, modello immenso". Below that, it lists "GLM-5.2 · 744B MoE · int4 · streaming CF". A code snippet shows a chat interaction with "colibrì v1.0 - GLM-5.2 · 744B MoE · int4 · streaming C" and Italian phrases like "pronto in 32s" and "Ciao! 😊 Come posso aiutarti oggi?". The section "The idea" details the model's architecture and performance.

    See all GitHub photosRead the GitHub wiki

    ?

    More GitHub photos

    See all GitHub photos
    rust-anything GitHub repositoryrust-anything GitHub repositoryAgent Substrate Google CNCF KubernetesAgent Substrate Google CNCF KubernetesFastMCP GitHub Repository v4.2FastMCP GitHub Repository v4.2BLA Blind LLM Agents GitHub repoBLA Blind LLM Agents GitHub repoAwesome-REA GitHub repositoryAwesome-REA GitHub repositoryWinnie the Pooh memeWinnie the Pooh memeOpen Ontologies GitHub releaseOpen Ontologies GitHub releaseSoMark OXR OCR benchmark resultsSoMark OXR OCR benchmark resultsHarness² self-improving AI agentsHarness² self-improving AI agentsOpenCode 2.0 releaseOpenCode 2.0 releaseREA Reverse Engineer Anything GitHubREA Reverse Engineer Anything GitHubREA Reverse Engineer Anything GitHub TrendingREA Reverse Engineer Anything GitHub Trendingopus-5.5 reverse-engineering toolopus-5.5 reverse-engineering toolGitHub streak patternGitHub streak patternMarvel Spider-Man SpidyVR modMarvel Spider-Man SpidyVR mod
    Photo
    Nate
    Nate@nate_5123mo
    📱GitHub💭Tech💭AI
    COLIBRI GLM-5.2 744B model

    @nate_512COLIBRI is wild. GLM-5.2, a 744-billion-parameter model, runs on a 25GB setup with zero GPU. Active params sit in RAM and the rest streams from disk on demand. It's not quick, but the responses are near-perfect on consumer hardware. Apache-2.0 and free open source, already at 24k stars on GitHub.

    View original post

    COLIBRI GLM-5.2 744B model

    Photo by @nate_512· Aug 12, 2026· GitHub

    About this photo

    The image focuses on a GitHub repository page for a project called "colibri". The page displays text describing a "Tiny engine, immense model" and its technical specifications. The mood is informative and technical, typical of software development documentation. A notable visual element is the pixelated bird logo for "colibri" in the upper left. The on-screen text includes "github.com" in the browser tab, and the project title "colibri" with Italian subtitles "piccolo motore, modello immenso". Below that, it lists "GLM-5.2 · 744B MoE · int4 · streaming CF". A code snippet shows a chat interaction with "colibrì v1.0 - GLM-5.2 · 744B MoE · int4 · streaming C" and Italian phrases like "pronto in 32s" and "Ciao! 😊 Come posso aiutarti oggi?". The section "The idea" details the model's architecture and performance.

    See all GitHub photosRead the GitHub wiki

    ?

    More GitHub photos

    See all GitHub photos
    rust-anything GitHub repositoryrust-anything GitHub repositoryAgent Substrate Google CNCF KubernetesAgent Substrate Google CNCF KubernetesFastMCP GitHub Repository v4.2FastMCP GitHub Repository v4.2BLA Blind LLM Agents GitHub repoBLA Blind LLM Agents GitHub repoAwesome-REA GitHub repositoryAwesome-REA GitHub repositoryWinnie the Pooh memeWinnie the Pooh memeOpen Ontologies GitHub releaseOpen Ontologies GitHub releaseSoMark OXR OCR benchmark resultsSoMark OXR OCR benchmark resultsHarness² self-improving AI agentsHarness² self-improving AI agentsOpenCode 2.0 releaseOpenCode 2.0 releaseREA Reverse Engineer Anything GitHubREA Reverse Engineer Anything GitHubREA Reverse Engineer Anything GitHub TrendingREA Reverse Engineer Anything GitHub Trendingopus-5.5 reverse-engineering toolopus-5.5 reverse-engineering toolGitHub streak patternGitHub streak patternMarvel Spider-Man SpidyVR modMarvel Spider-Man SpidyVR mod