Top all-time · #7497
Morph
Company blog · morphllm.com/blogaiprogramming
Morph is the company blog at morphllm.com, writing about ai and programming. Since 2025, 11 different Hacker News readers have linked it 21 times. No feed was found, so read it on the site itself.
Too few sampled comments (7) to say how often HN recommends it.?
Why HN links it
Morph's site covers AI code-editing tools for coding agents, including fast-apply models, a compaction product and a playground. HN comments mostly ask about or discuss the product rather than the writing.
Summarised by Claude from a sample of 7 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
GLM-5.3 with 1M context on inference tuned for coding agents. Plus the agent toolkit: search (WarpGrep), edits (Fast Apply), context (Compact). One OpenAI-compatible API.
Most-linked posts?
- /products/fastapply2 people · 2 mentions · first linked 2026-02-25
- /dashboard/playground/apply2 people · 2 mentions · first linked 2025-07-08
What Hacker News says?
Insane work. For sure the middle ground between Figma and Cursor thats been missing. Happy to be powering Fast Applies in Onlook! https://morphllm.com
Comment linking /
on “Show HN: Onlook – Open-source, visual-first Cursor for designers”Submitted Optimizing Models to Be Fast at Codegen as a story · 2 points
HN discussionit really matters when you care deeply about the product experience. The gap between something that works sometimes and something that works great everytime is what separated Cursor from the rest. Try it out: https://morphllm.com/dashboard/playground/apply
Comment linking /dashboard/playground/apply
on “Launch HN: Morph (YC S23) – Apply AI code edits at 4,500 tokens/sec”Submitted 10,500 tokens/SEC per request on Nvidia hardware as a story · 2 points
HN discussionmcp is out! https://morphllm.com/mcp
Comment linking /mcp
on “Launch HN: Morph (YC S23) – Apply AI code edits at 4,500 tokens/sec”Submitted Show HN: Learning to Test Code Changes with RL as a story · 1 points
HN discussionmcp is out! https://morphllm.com/mcp
Comment linking /mcp
on “Launch HN: Morph (YC S23) – Apply AI code edits at 4,500 tokens/sec”Submitted Show HN: Parallel Agentic Search on the Twitter Algorithm as a story · 1 points
HN discussionSeems completely broken. I used the provided HTML example on https://morphllm.com/dashboard/playground/apply . Without editing anything at all, I pressed apply. Your model added a bunch of CSS even though that wasn't in the update instructions at all. It also added a contact section, which again, wasn't in the update instructions that your demo provided.
Comment linking /dashboard/playground/apply
on “Launch HN: Morph (YC S23) – Apply AI code edits at 4,500 tokens/sec”Submitted WarpGrep – RL Subagent for Fast Context (Like SWE-Grep) as a story · 1 points
HN discussionReally like this. I've been trying microsoft's copilot and it's so clunky, particularly when applying edits. One would assume they have the resources to train the model.. Request: please provide a system prompt in the docs to help the llm generate the diff format that performs best w/ your models. LLMs frequently change the way they present diffs on upgrades and I don't want to be guessing which format is best. EDIT…
Comment linking /privacy
on “Launch HN: Morph (YC S23) – Apply AI code edits at 4,500 tokens/sec”Submitted Everything Is Model[S] as a story · 1 points
HN discussionIts hard to get a model that does it usefully on your laptop. Theres an open source 1.5B model from QuocDat, and Morph - https://morphllm.com which is a fast apply model as an API (that I run)
Comment linking /
on “Reverse Engineering Cursor's LLM Client”Submitted MCP for Fast AI Filesystem Edits as a story · 1 points
HN discussion> Sonnet/Claude Code may technically be "smarter", but Qwen3-Coder on Cerebras is often more productive for me because it's just so incredibly fast. Saying "technically" is really underselling the difference in intelligence in my opinion. Claude and Gemini are much, much smarter and I trust them to produce better code, but you honestly can't deny the excellent value that Qwen-3, the inference speed and $50/month for…
Comment linking /
on “Cerebras systems raises $1.1B Series G”Submitted LLMs Are Bad at Being Forced as a story · 1 points
HN discussionHow do you compare against Morph ( https://morphllm.com/ )?
Comment linking /
on “Launch HN: Relace (YC W23) – Models for fast and reliable codegen”check out this: https://www.morphllm.com/products/compact you can wire it into pi compaction pretty easily
Comment linking /products/compact
on “How Compaction Works in Pi”https://morphllm.com/benchmarks Using fast apply is more reliable at first pass and is faster/cheaper. You prompt your agent to output in a lazy format and our model learns hoe to merge it in. The ides listed typically do turn based search and replace which uses more tokens and takes longer Kilo supports morph as well!
Comment linking /benchmarks
on “10,500 tokens/SEC per request on Nvidia hardware”In the same realm to compare to https://www.morphllm.com/products/fastapply
Comment linking /products/fastapply
on “Mouse: Precision Editing Tools for AI Coding Agents”I changed my mind: this would be perfect for a fast edit model ala Morph Fast Apply https://www.morphllm.com/products/fastapply It looks like they are offering this in the form of "Mercury Edit"and I'm keen to try it
Comment linking /products/fastapply
on “Mercury 2: The fastest reasoning LLM, powered by diffusion”