Miru: Code Search Built for AI Agents改善

Hybrid code search that finds code by meaning, not pattern-matching — so agents stop burning tokens reading the wrong files.

Hybrid code search that finds code by meaning, not pattern-matching — so agents stop burning tokens reading the wrong files.

THE CHALLENGE

AI coding agents don’t spend most of their tokens writing code — they spend it looking for it.

Grep and glob are cheap per call but expensive in aggregate — an agent using them reads far more of the repo than it needs, and that waste compounds with every retry. Miru targets the ~85% of the token budget that conventional search leaves untouched.

Measured across real tasks, the token bill breaks down like this:

Where do the tokens go?

HOW IT WORKS

Hybrid retrieval, chunked on meaningful boundaries

Miru combines semantic search — via ds1-code, the code-specialised model in the DS1 embedding family — with classic lexical search (BM25), then fuses and reranks the results. Embeddings capture intent; BM25 nails exact tokens like a symbol name. And because ds1-code runs on CPUs we have no limits on usage.

How it works

WHAT IT DELIVERS — MEASURED

Miru vs. grep, end to end, across 7 coding scenarios

Grep looks cheap per call, but the agent compensates by reading more, retrying more, and writing more. Those costs only show up when you measure the whole task.

Hybrid code search that finds code by meaning, not pattern-matching — so agents stop burning tokens reading the wrong files.

In Takara’s own benchmark, an agent equipped with Miru was run end-to-end against the same agent using grep, across seven coding scenarios. Miru came out ahead on every dimension that matters:

• Up to 50% cheaper, 20% cheaper on average, per question answered

• Up to 60% faster, 11% faster on average, wall-clock per question

• Up to 45% fewer steps, 17% fewer on average, for the agent to reach an answer

WHERE MIRU FITS

Built for agents, teams, and regulated codebases alike

Where Miru fits

DEPLOYMENT, PRIVACY & SOVEREIGNTY

CPU-only, Cloud or on-premise, we have you covered.

Because ds1-code is CPU-only, Miru's retrieval needs no scarce GPUs and can run anywhere standard compute is available. The open-source MCP server embeds through Takara's hosted API by default; for sensitive code, the same architecture supports cloud based deployment in your environment — code never leaves your perimeter.

GETTING STARTED

Three commands to wire Miru into your agent

bun add -g @takara-ai/miru-code # install

miru setup # store your Takara API key

miru install # wire into your IDE / agent

miru search "auth middleware" ./src