# Morph

> Fast, specialized inference for coding agents: Fast Apply code editing, WarpGrep agentic code search, Compact context compression, Reflex classifiers, and fast open-weight coding models — one OpenAI- and Anthropic-compatible API at https://api.morphllm.com.

Every page below is also served as markdown: append `.md` to the URL, or request it with `Accept: text/markdown`.

## Product
- [Home](https://www.morphllm.com/index.md): What Morph is: fast, specialized inference for coding agents
- [/pricing](https://www.morphllm.com/pricing.md): Prices for every model and API
- [/dedicated-inference](https://www.morphllm.com/dedicated-inference.md): Dedicated LLM endpoints with reserved B200 or B300 capacity
- [/dedicated-inference/calculator](https://www.morphllm.com/dedicated-inference/calculator.md): LLM inference cost, capacity, and coding agent fleet calculator
- [/benchmarks/dedicated-inference](https://www.morphllm.com/benchmarks/dedicated-inference.md): Plain language B200 and B300 inference benchmark analysis
- [/glm-5-3-flash](https://www.morphllm.com/glm-5-3-flash.md): GLM 5.3 Flash API, pricing, use cases, and dedicated capacity planning
- [/models](https://www.morphllm.com/models.md): Fast general coding models (GLM-5.3, Kimi K3, DeepSeek V4 Flash)
- [/products/fastapply](https://www.morphllm.com/products/fastapply.md): Fast Apply: merge LLM code edits at 10,500+ tok/s
- [/products/warpgrep](https://www.morphllm.com/products/warpgrep.md): WarpGrep: agentic code search subagent
- [/products/compact](https://www.morphllm.com/products/compact.md): Compact: context compression
- [/products/reflex](https://www.morphllm.com/products/reflex.md): Reflex: per-turn classifiers
- [/products/glance](https://www.morphllm.com/products/glance.md): Glance: AI browser testing on PR previews
- [/changelog](https://www.morphllm.com/changelog.md): Product changelog
- [/benchmarks](https://www.morphllm.com/benchmarks.md): Benchmarks

## API
- [OpenAPI description](https://api.morphllm.com/openapi.json): machine-readable spec for every endpoint
- [API catalog](https://api.morphllm.com/.well-known/api-catalog): RFC 9727 discovery document
- [Live model list + prices](https://www.morphllm.com/api/models/json): fetch this rather than hardcoding model facts

## Docs
- [Documentation](https://docs.morphllm.com): guides and API reference (append `.md` to any docs URL for markdown)
- [Docs llms.txt](https://docs.morphllm.com/llms.txt): docs-scoped index

## Writing
- [/blog/agent-failures-dont-throw](https://www.morphllm.com/blog/agent-failures-dont-throw.md)
- [/blog/all-agents-coding-agents](https://www.morphllm.com/blog/all-agents-coding-agents.md)
- [/blog/best-practices](https://www.morphllm.com/blog/best-practices.md)
- [/blog/bitter-lesson](https://www.morphllm.com/blog/bitter-lesson.md)
- [/blog/claude-code-cost](https://www.morphllm.com/blog/claude-code-cost.md)
- [/blog/claude-code-mcp-servers](https://www.morphllm.com/blog/claude-code-mcp-servers.md)
- [/blog/code-search-bottleneck](https://www.morphllm.com/blog/code-search-bottleneck.md)
- [/blog/codegen-inference-research](https://www.morphllm.com/blog/codegen-inference-research.md)
- [/blog/coding-agent-harness-lessons](https://www.morphllm.com/blog/coding-agent-harness-lessons.md)
- [/blog/compact-sdk](https://www.morphllm.com/blog/compact-sdk.md)
- [/blog/compute-scarcity-kernels](https://www.morphllm.com/blog/compute-scarcity-kernels.md)
- [/blog/cursor-apply](https://www.morphllm.com/blog/cursor-apply.md)
- [/blog/cursor-mcps](https://www.morphllm.com/blog/cursor-mcps.md)
- [/blog/deepseek-v4-1-flash](https://www.morphllm.com/blog/deepseek-v4-1-flash.md)
- [/blog/diffs-vs-fast-apply](https://www.morphllm.com/blog/diffs-vs-fast-apply.md)
- [/blog/everything-is-models](https://www.morphllm.com/blog/everything-is-models.md)
- [/blog/fast-apply-fast-agents](https://www.morphllm.com/blog/fast-apply-fast-agents.md)
- [/blog/fast-context-rl-retrieval](https://www.morphllm.com/blog/fast-context-rl-retrieval.md)
- [/blog/faster-agents](https://www.morphllm.com/blog/faster-agents.md)
- [/blog/glm-5-2](https://www.morphllm.com/blog/glm-5-2.md)
- [/blog/llms-bad-at-being-forced](https://www.morphllm.com/blog/llms-bad-at-being-forced.md)
- [/blog/long-running-agents](https://www.morphllm.com/blog/long-running-agents.md)
- [/blog/morph-aws-case-study](https://www.morphllm.com/blog/morph-aws-case-study.md)
- [/blog/morph-breaks-10k-barrier](https://www.morphllm.com/blog/morph-breaks-10k-barrier.md)
- [/blog/morph-continue](https://www.morphllm.com/blog/morph-continue.md)
- [/blog/morph-decode](https://www.morphllm.com/blog/morph-decode.md)
- [/blog/morph-gets-faster](https://www.morphllm.com/blog/morph-gets-faster.md)
- [/blog/multi-agent-systems](https://www.morphllm.com/blog/multi-agent-systems.md)
- [/blog/reflex-inference-engine](https://www.morphllm.com/blog/reflex-inference-engine.md)
- [/blog/self-hosting](https://www.morphllm.com/blog/self-hosting.md)
- [/blog/thinking-fast-and-slow](https://www.morphllm.com/blog/thinking-fast-and-slow.md)
- [/blog/warpgrep-v2](https://www.morphllm.com/blog/warpgrep-v2.md)
- [/blog/what-is-morph-for](https://www.morphllm.com/blog/what-is-morph-for.md)

## Optional
- [Full concatenated corpus](https://www.morphllm.com/llms-full.txt): everything above in one document
