Tweet ghim
Morph
406 posts
Morph
@morphllm
ultra fast specialized models that improve coding agents https://t.co/dBQjovGya3 Try WarpGrep: https://t.co/dXjJCKwINV
San Francisco, CA Tham gia Mayıs 2025
3 Đang theo dõi3K Người theo dõi
@jackjoliet you can specify that in the sdk if you use the messages input
docs.morphllm.com/sdk/components…
English
We just removed access control. Anyone can now try this from the API as well.
docs.morphllm.com/sdk/components…
English
Morph đã retweet
Perfect compaction is a prerequisite for long-running agents.
It’s the difference between a country of geniuses
and a pile of clankers.
#unLobotomizeClaude
Morph@morphllm
Introducing FlashCompact - the first specialized model for context compaction 33k tokens/sec 200k → 50k in ~1.5s Fast, high quality compaction
English
@scent_fetish_ Not in the mcp yet - just npm for now. For opencode we have a plugin github.com/morphllm/openc…
English
Those guys are doing it right.
@morphllm
Do we update MCPs, or do they just work server-side (NPM use, not NPX)?
Morph@morphllm
Introducing FlashCompact - the first specialized model for context compaction 33k tokens/sec 200k → 50k in ~1.5s Fast, high quality compaction
English
English
@itspers @albfresco @mots_pod How could we make it less work?
playground here morphllm.com/dashboard/play…
English
@morphllm @albfresco @mots_pod Too much work to try. Would be cool if it was available on open router.
English
@peteyburn @albfresco @dhruvbhatia0 @mots_pod there are hooks for it but we're looking into it. There's no (public) interface for modifying context
English
@albfresco @dhruvbhatia0 @morphllm @mots_pod But can you force Claude to clear it’s context with Morph’s output?
English
@morphllm context-free compression is the dream. 200k to 50k in 1.5s is actually insane throughput.
English
@turbo_xo_ @sama for what its worth the codex /compactions endpoint is pretty decent. Unfortunately encrypted outputs and a bit slow though
English