BitRouter

BitRouter

The self-improving LLM router that optimize your agentic workflows with every runs, works with any harnesses, any models, any loops.

You're tokenmaxxing in production. Every step of every loop bills at frontier prices — file reads, tool calls, sub-agent hops, retries. Most don't need it. BitRouter routes each call, tool, and agent to the cheapest path that still reaches the goal, and tightens that routing as the loop runs.

Cost is live today — latency and accuracy are next.

Three primitives, one gateway

An agentic loop consumes three things. Other routers govern only the first. BitRouter makes all three routable, observable, and governed:

• Models — route LLM calls across providers, accounts, and wire protocols: OpenAI Chat Completions, OpenAI Responses, Anthropic Messages, and Google Gemini. (the classic router, cross-protocol — any request format to any upstream, and back)
• Capabilities — an MCP gateway and an AgentSkills gateway: tools and skills become governed, routable resources instead of hardcoded endpoints. (The skills gateway folds into the MCP gateway once the MCP skills extension reaches production.)
• Agents — an ACP gateway: sub-agents become first-class routable primitives, so a task can go to the sub-agent that best fits the loop's objective — just as a call routes to the best-fit model. (Local sub-agents over stdio today; remote gateways arrive with ACP v2.)

Optimizing a loop isn't just model selection — it's choosing the model, the tool, and the sub-agent that best serve the loop's objective at every step that gets it to its goal.

The self-improving loop

BitRouter wraps your agentic loop in a second loop. Each loop gets its own policy spec — a config file that declares how its calls, tools, and agents should route. Against that spec BitRouter runs a continuous act → observe → evaluate → learn cycle, and every step is a component it already ships:

• Act — the router. Each model, tool, and agent call is rewritten to a chosen route: policy-table routing, cross-protocol translation, multi-account failover.
• Observe — telemetry. Every hop is attributed with cost, tokens, latency, and outcome, and exported to Prometheus or any OTLP backend.
• Evaluate — the eval engine. Each run and routing decision is scored against your chosen objective — did the route it picked still reach the goal?
• Learn — the policy engine. The eval signal folds back into the policy spec: an agent self-tunes it, or you edit it by hand. The next turn of the loop acts on the improved spec.

You choose what the loop optimizes for — cost, latency, or accuracy — and it improves the longer it runs in production.

Benchmarks

Today cost is the validated objective: on Terminal-Bench 2.1, gpt-5.5 with BitRouter cut cost 32.8% at near-parity accuracy (−1.1 pp), by offloading routine steps to a cheaper model. Latency and accuracy objectives — and more base models — are landing next.

Base model	Cost vs baseline	Latency vs baseline	Accuracy vs baseline
gpt-5.5	**−32.8%**¹	coming soon	coming soon
gpt-5.6	coming soon	coming soon	coming soon
claude-opus-4.8	coming soon	coming soon	coming soon
claude-sonnet-5	coming soon	coming soon	coming soon
claude-fable-5	coming soon	coming soon	coming soon

¹ Cost-optimization run on Terminal-Bench 2.1: −32.8% zero-cache imputed cost (audited range 28.6–32.8% by cache share) at near-parity accuracy, −1.1 pp (76.1% vs 77.3%, within single-attempt noise).