OpenAI Deep Research: how many tokens per task, and what does it cost?
OpenAI Deep Research publishes no per-task cost or token figure in the AgentBurn index. AgentBurn snapshot 2026-10-27. Every figure carries a provenance label: source-backed, derived, or left empty when no public basis exists.
What it is. Long-horizon research agent that browses and writes a cited report
Deep Research plans a multi-step investigation, browses dozens of pages, cross-checks sources and returns a structured report with citations in 5-30 minutes. It is priced as a premium ChatGPT tier or per-query API spend, which puts a single deep query at several dollars of inference.
| Tokens per task | not published | Not published |
|---|---|---|
| Cost per task | not published | Not published |
| Success rate | not published | Not published |
| Cost per solved task | not published | — |
| Tool calls per task | not published | — |
| Retry overhead | not published | — |
| Minutes per task | not published | — |
| Estimated runs per day | null (model estimate) | Not published |
Where the tokens go
This is the per-task split AgentBurn records for OpenAI Deep Research, step by step.
| Step | What happens | Share of tokens | Tool calls |
|---|---|---|---|
| Clarify scope | Restates the question and asks follow-ups | 6% | 2 |
| Plan searches | Plans the search tree and source types | 10% | 4 |
| Browse | Reads pages, PDFs and datasets, extracts evidence | 36% | 22 |
| Cross-check | Verifies claims against multiple sources | 24% | 10 |
| Write report | Produces the cited long-form report | 24% | 7 |
What each number is based on
AgentBurn labels every field. Source-backed means the cited page states the figure; derived means a formula on this page produced it; not published means no public basis exists and the field is left empty rather than estimated.
| Field | Basis | Note |
|---|---|---|
| tokensPerTask | Not published | No usable public source for per-task tokens. |
| costPerTaskUsd | Not published | No usable public source for per-task cost. |
| successRatePct | Not published | No public task success rate. |
| runsPerDay | Not published | No public run volume. |
Frequently asked
How many tokens does OpenAI Deep Research use per task?
No public tokens-per-task figure exists for OpenAI Deep Research; AgentBurn leaves the field empty rather than estimating it.
What does OpenAI Deep Research cost per task?
No public per-task cost exists for OpenAI Deep Research, so it is recorded as not published.
Is this a vendor list price?
No. It is a per-task figure from a cited source, not a rate card; the two are not comparable.
Sources
- OpenAI — Introducing deep research (product documentation)
- Vendor site: https://openai.com/index/introducing-deep-research/
- Open dataset: /data/projects.json
- Methodology: how these numbers are built
Cite this page
AgentBurn (2026-10-27). "OpenAI Deep Research". https://getagentburn.com/projects/deep-research/
Machine-readable: projects.json · Markdown version