Morph
Verified Oct 15, 2026 ActiveAPI of specialized models for coding agents—fast code-apply, code search and context compression—priced per million tokens.
Listed on EffectHub ·
Pricing
- Pricing model
- Freemium
Pricing is usage-based per million tokens (e.g. Fast Apply 7B at $0.80 input / $1.20 output) rather than a flat plan price; a free tier covers 200 requests a month plus $10 of compute.
Use tool ↗About Morph
Morph doesn't build a coding agent itself; it sells the specialized models other coding agents call for specific subtasks, on the theory that a general-purpose LLM is slower and pricier than a model tuned for one job. Its lineup includes Fast Apply models for editing code, a search model called WarpGrep, a context-compression model, embeddings and reranking, all billed per million tokens through an OpenAI-compatible API. It plugs into coding tools via an MCP server for Cursor, Claude Code, Windsurf, Cline, VS Code and Claude Desktop, plus a Vercel AI SDK integration. Morph is a Y Combinator-backed company based in San Francisco; pricing is entirely usage-based with a free tier of 200 requests a month.
- Platforms
- API
- API
- Yes
- Last verified
- Oct 15, 2026
Verified sources
- Product lineup: Fast Apply, WarpGrep, Compact, embeddings, rerank, with per-token pricingmorphllm.com ↗
- MCP server for Cursor, Claude Code, Windsurf, Cline, VS Code and Claude Desktopmorphllm.com ↗
Is Morph legit?
We don’t label tools safe or unsafe. Below are the records we checked ourselves, each with the date we read it.
- Domain registered
- Feb 6, 2025 (1 year ago)Registry record (RDAP), registrar NameCheap, Inc., looked up Sep 6, 2026
- TLS certificate
- Issued by Let's Encrypt, valid until Jan 13, 2027Seen in our own connection to the site on Oct 15, 2026
- Last checked
- Oct 15, 2026The site answered
- Operated by
- Morph, United StatesOur research, card verified Sep 1, 2026
Pros & cons of Morph
Morph's bet is that coding agents get faster and cheaper by routing specific subtasks—applying an edit, searching a codebase, compressing context—to small purpose-built models instead of a general LLM for everything. That's a real architectural choice with measurable cost implications, but it only pays off if you're already building or customizing an agent; it's infrastructure, not a product an end user would touch directly.
Pros
- Purpose-built models for specific agent subtasks instead of one general model for everything
- OpenAI-compatible API plus an MCP server, so integration into existing tools is straightforward
- Free tier and per-token pricing published openly
Cons
- Usage-based pricing makes total monthly cost hard to predict in advance
- Only useful if you're already building or configuring a coding agent, not a standalone tool
Best for
- Teams building custom coding agents who want to cut general-LLM costs on repetitive subtasks
- Developers already using Cursor, Claude Code, Windsurf or Cline via MCP
- Engineers optimizing agent latency on code-editing and search steps specifically
Not for
- Anyone wanting a finished, end-user coding assistant rather than an API layer
- Teams without the engineering time to integrate a second model provider
Alternatives to Morph
Morph head-to-head
Side-by-side fact sheets against the competitors our research or their makers name — pricing, platforms, API and both verdicts on one page.