Seer bills your team. We bill your usage.
How much does Sentry Seer cost?
Sentry Seer costs $40 per active contributor per month, on top of your Sentry plan. Sentry defines an active contributor as any user who makes 2 or more pull requests to a seer-enabled repository. all repository members except github bots marked [bot] are counted. The bill is therefore a function of how many people commit to the repository, not of how many errors the agent fixes.
The previous model — $20 per month per subscription plus $25 of Seer event credits — stopped being offered as an add-on in January 2026. Source: Sentry pricing docs, read July 2026.
What that means per team size
| Active contributors | Seer, per month | Cost of fixing nothing |
|---|---|---|
| 3 | $120 | $120 |
| 8 | $320 | $320 |
| 20 | $800 | $800 |
| 50 | $2000 | $2000 |
The third column is the point. Seer's seat fee does not care whether the agent ran. Parsemend's does: no fix run means no tokens, and the framework fee is flat.
| Capability | Parsemend | Sentry Seer |
|---|---|---|
| What you pay for | ||
| Priced on | Flat fee + your token spend | $40 / active contributor / mo |
| Bill grows with team size | No — grows with usage, not headcount | Yes — every contributor counts |
| Inference resold with a margin | No — your provider bills you directly | Not documented |
| Costs nothing when nobody fixes anything | Yes — no run, no tokens | No — the seat fee is flat |
| Control | ||
| Choose the model | Yes | Not documented |
| Bring your own API key | Yes | Not documented |
| Run it entirely on your own hardware | Yes — CLI, or your own agent, with Ollama | No — analysis runs in Sentry's cloud |
| Works without repository access | Yes — in three of the four modes | No — GitHub or GitLab app required |
Why we do not resell you inference
If a vendor marks up your tokens, it earns more when the agent thinks harder. We would rather not hold that incentive. You bring the API key, your provider bills you at their public rate, and Parsemend charges a flat fee for the framework around the model — context bundles, the resumable state machine, confidence gating, draft PR post-back.
todo— Parsemend's own tier prices are not final, so this page deliberately compares the shape of the two bills rather than printing a head-to-head total we cannot stand behind yet.
A flat seat price is easy to forecast, and Seer's is bundled into a platform your team probably already runs. Two bills and an API key are more work. If your team is three people and nobody wants to think about tokens, Seer at $120 a month is a perfectly sensible purchase.
Frequently asked questions
How much does Sentry Seer cost?
As of July 2026, Seer costs $40 per active contributor per month, added to your existing Sentry subscription. An active contributor is any user who makes two or more pull requests to a Seer-enabled repository; every repository member except GitHub bots is counted. The earlier model, $20 per month plus $25 of event credits, stopped being offered in January 2026.
Is Seer priced per fix?
No. Seer is priced per active contributor per month, so the bill tracks the size of your team rather than how many issues you actually fix. A ten-person team pays $400 a month whether the agent opens one pull request or a hundred.
Can I use my own model or API key with Seer?
Sentry does not document a way to choose the model or supply your own API key. Parsemend supports Anthropic, OpenAI, Gemini and self-hosted Ollama, and in three of its four modes the inference is billed to your provider account rather than to Parsemend.
Does Seer run on my infrastructure?
No. Seer analyses issues in Sentry's cloud. The local-development integration announced in January 2026 is an MCP server: telemetry still flows to Sentry, where the analysis happens. Parsemend's CLI mode runs the entire fix loop on your machine.
See also Parsemend vs Sentry and the four fix modes.