See where agent spend goes
Compare session cost by output type before you change how a team works.
$246 · 11 sessions
- pr · 6$171
- doc · 2$12
- research · 1$18
- ops · 1$21
- nothing · 1$24
$22 average per session · PR sessions 70% of spend

Pensero Agent Traffic Control
Measure the cost per delivered point for each owner. Pensero keeps premium models where they deliver and routes the others to lower-cost models.
Compare each owner’s agent spend with the delivery points their sessions produced, in USD per point.
Owners within target pass unchanged to the model they asked for. Low delivery is flagged, never blocked.
When an owner’s $/pt sits above target, premium requests go to a lower-cost model at your provider, with reasoning effort capped.
Compare session cost by output type before you change how a team works.
$246 · 11 sessions
$22 average per session · PR sessions 70% of spend

Compare cost per point per owner to decide where an expensive model is worth keeping.
$20 / pt default target
1 of 3 owners within the $20 / pt target

Delivery points multiply the magnitude of each PR by its complexity. Our proprietary technology combines AI, heuristics and ML to give deterministic results.
6 scored PR sessions · 14 points
2.3 points per scored PR session on average

Route premium requests to a lower-cost model; investigate low delivery without blocking models.
Policy applied to each request
Last 14 days · target $20 / pt

Compare the cost of token usage with the tasks that sessions delivered. Use USD per delivery point ($/pt) for each owner over a configurable 14-day window.
Every request carries the owner it belongs to and the model it asked for.
Fewer than 3 scored sessions, or no rating? The request passes unchanged.
$/pt against your target. Low delivery passes, flagged only.
Yes: fall back to a lower-cost model. Otherwise, pass.
Your session data stays on your host. Send agent requests to any OpenAI-compatible endpoint or LiteLLM provider, including OpenAI, Gemini, vLLM and Ollama, or Bedrock or Anthropic.
Proxy, scorer, session database and dashboard run on your host. Session data never leaves it.
Only agent requests and their context go out, to any OpenAI-compatible endpoint or LiteLLM provider.
One Docker Compose deployment, a proxy key and a password-protected dashboard. Add a local PR scoring model if you want.
Agent spend grows fast, and a token bill does not show what the work delivered. Pensero puts the cost of each session next to the work it produced, so you can see which usage pays off.
Pensero scores Claude Code and Codex CLI sessions today. Agents can use any model provider, such as OpenAI, Anthropic, Gemini, Bedrock or a local model.
They are a proxy for it. Pensero measures the magnitude and difficulty of the work that each session delivers. Use them together with a review of the work, not to rank people.
Pensero compares each owner's spend with the work they delivered. Where a premium model delivers, it stays. Where the cost per point is too high, requests go to a lower-cost model. Low delivery gets a flag, not a block. You set the thresholds.
Pensero runs on your host or VPC. Session data, scores and the dashboard stay there. Model requests go only to the providers that you choose, and these can be local.
Book a call. We deploy it on-prem, on your host or VPC.