โธ OPS-BRIEF / SECTOR: DEV-INFRA
h.HotCopy.ai
Your codebase is too big for any context window. So we killed the context window. Managed recursive AI coding CLI โ agent swarms decompose impossible problems in parallel.
โธ UNBOUNDED CONTEXT // 50+ PARALLEL AGENTS // 0 API KEYS
## What HotCopy is
Managed recursive AI coding CLI. Agent swarms decompose impossible problems in parallel. No context limits. No API keys. No config. Just install and go. HotCopy treats your project as an *environment* the model interacts with programmatically โ never raw files in a prompt.
## The Recursive Language Model paradigm
HotCopy implements the Recursive Language Model (RLM) paradigm introduced by Zhang, Kraska, and Khattab at MIT CSAIL ([arXiv:2512.24601](https://arxiv.org/abs/2512.24601), Jan 2026). Existing scaffolds (Claude Code, Gemini CLI, Cursor) place the user prompt directly into the LLM's context window, generate output autoregressively (capped by token limits), and verbalize sub-LLM calls (only O(1) delegations). RLMs invert this: the prompt is loaded as a REPL variable, the model writes code to interrogate it, sub-calls are programmatic functions inside that code, and the answer is built up in variables and returned via `FINAL_VAR()` โ unbounded by any single model's generation length.
## Benchmarks vs base model
| Benchmark | Input Size | Base GPT-5 | RLM(GPT-5) | Avg Cost |
|---|---|---|---|---|
| BrowseComp+ | 6โ11M tokens | **0%** | **91.3%** | $0.99 |
| OOLONG | 131K tokens | 44% | 56.5% | $0.43 |
| OOLONG-Pairs | 32K tokens | 0.04% | 58.0% | $0.33 |
| CodeQA | 23Kโ4.2M tokens | 24% | 62% | $0.11 |
Median RLM cost โค base model cost; up to 3ร cheaper than summary-agent baselines. Source: Zhang, Kraska, Khattab. *Recursive Language Models.* MIT CSAIL. [arXiv:2512.24601](https://arxiv.org/abs/2512.24601), Jan 2026.
## How it works
- **Prompt as environment.** Project context is loaded into Durable Object memory as REPL variables. The orchestrator sees only metadata (file count, repo map, total chars) โ never raw files.
- **Two-tier orchestrator/worker architecture.** The RLM root model owns a 256K orchestration window. Every sub-call routes to a separate worker model with its own 256K window and multimodal capability โ managed for you, no provider keys to wire up.
- **REPL with stdout truncation.** The orchestrator writes JavaScript that runs in a sandboxed V8 isolate (`@cloudflare/codemode`). Stdout is truncated before being appended to history โ variables hold the long state, the model history holds the metadata.
- **Parallel sub-calls via `llm_batch()`.** Unlike the paper's sequential implementation, HotCopy fans out scouts in parallel through Cloudflare Workers, dramatically reducing latency on map-reduce decompositions.
- **Durable Objects + AI Gateway.** Every model call routes through `hotcopy-gateway` for cost tracking, prefix caching (per-session affinity), and rate-limit handling. State lives in DO SQL โ trajectories, sub-calls, costs, all logged for debugging.
- **`FINAL_VAR()` for unbounded output.** The model returns the *name* of the variable holding the answer. The reply is whatever that variable holds โ no autoregressive output cap.
## Architecture deep-dive