Commercial intent · measure completed-task economics
Calculate Claude Opus 5 cost for Unreal coding and agent tasks using official $5/$25 token pricing, effort, cache behavior, tools, retries, review time, and task-level budgets.
Anthropic launched Opus 5 at $5 per million input tokens and $25 per million output tokens, the same base API rate as Opus 4.8. Fast mode is listed at roughly 2.5× speed and twice the base price. Unreal teams should budget by completed, accepted task—not prompt—because repository context, tool outputs, retries, failed builds, long-agent loops, and human review can dominate total cost.
Choose a concrete prompt, open the SEELE Unreal creator, review the prefilled brief, and generate the native project. The first prompt is tailored to this page; the remaining prompts are reusable starting points.
Create a native Unreal 5 delivery-race prototype in a compact alpine village. The player collects one parcel, chooses between a safe road and a faster rooftop route, reaches the recipient before a visible timer expires, and can immediately retry. Keep the scope to one polished route decision, clear feedback, keyboard and controller support, and a browser-preview session under four minutes.
Use this prompt in SEELEUse Unreal Engine to generate a simple third-person tour game with mountains and water. Include responsive movement, a clear route, one completion state, failure recovery, immediate restart, browser preview, and a native downloadable Unreal 5 project.
Use this prompt in SEELEBuild a compact dungeon crawler using Unreal Engine. Include third-person movement, one combat loop, two connected rooms, one key-and-door objective, visible health and feedback, death, completion, immediate restart, browser preview, and a native downloadable Unreal 5 project. Test and regress the result.
Use this prompt in SEELECreate a native Unreal 5 third-person city tour with a compact walkable district, vegetation, landmark lighting, and sunny, cloudy, and rainy weather states. Include clear controls, browser preview, stable restart behavior, packaging checks, and local project download.
Use this prompt in SEELEAn inspectable Unreal project rather than a model answer or an unofficial integration claim.
A fast way to check camera, controls, objective clarity, feedback, completion, failure, and restart before local continuation.
A workflow for performance review and package preparation before external distribution.
A local project handoff for source, Blueprint, asset, plugin, rights, build, target-device, and release review.
The July 24 launch lists $5 per million input tokens and $25 per million output tokens for claude-opus-5.
Anthropic says Fast mode runs around 2.5 times the default speed and costs twice the base price on the Claude Platform or through Claude Code usage credits.
The release highlights effort settings as a way to trade intelligence, tokens, speed, and cost; the best setting depends on the task and acceptance rate.
Context preparation, tool execution, build minutes, retries, rejected edits, reviewer time, and delayed feedback can cost more than the token line item.
Use an accepted Unreal task: a reviewed plan, merged small diff, correctly diagnosed issue, passing test, or approved design brief. Counting prompts rewards retries and hides failed work.
Record uncached and cached input, output, thinking or effort setting, tool calls, subagents, retries, fallbacks, elapsed time, build and test compute, failures, and reviewer minutes for each task.
Replay the same known C++, Blueprint, log, planning, and long-agent tasks against the current model and at least one cheaper candidate. Keep permissions, context, acceptance checks, and target environment identical.
Reserve Opus 5 for tasks where its accepted-result rate or reduced review effort offsets the price. Stop loops with no new evidence, repeated compilation failures, expanding scope, or cost beyond the task ceiling.
If the user wants a finished game project, compare the model-assistance budget with using SEELE AI to generate, preview, optimize, package, and download the native Unreal project. These are different products and cost units.
Quoting only per-token rates can make a high-retry agent look cheaper than it is.
Large repository dumps increase input cost and privacy risk without guaranteeing useful context.
Fast mode may reduce elapsed time but increase spend without improving accepted results.
Pricing, plan access, credits, cache rules, limits, and regional availability can change after the cited launch date.
| Dimension | What good looks like | Evidence to keep |
|---|---|---|
| Token spend | Input and output cost for the complete task | Provider usage record tied to task ID |
| Tool spend | Build, test, computer-use, network, and infrastructure cost | Tool logs and compute duration |
| Human spend | Preparation, review, correction, and handoff time | Time sample from the same task class |
| Value | Accepted tasks, escaped defects, cycle time, and reusable evidence | Compare matched baseline over a meaningful sample |
Combine input and output tokens, cache behavior, tool calls, build minutes, retries, failed runs, context preparation, reviewer time, test infrastructure, and escaped defects. The lowest API line item can still produce the highest cost per accepted Unreal change.
Set separate ceilings for repository questions, small code reviews, root-cause investigations, Blueprint planning, long-running feature work, computer use, and release checks. Each class has different context, tool, latency, retry, and evidence requirements.
When the business goal is a playable game, measure time to a coherent native project and browser preview, then the effort to review, download, optimize, package, and ship. Do not confuse model-token economics with end-to-end game-production economics.
Use the model research to tighten the brief, not to replace project evidence. The direct creation path is deliberately short and observable:
Keep one camera, one primary verb, one objective chain, explicit failure and completion, restart behavior, visual direction, controls, target session length, and a clear cut list.
Open the canonical Unreal creator, choose the closest verified starter world, submit the brief, and generate a native Unreal 5 project rather than treating a text answer as the deliverable.
Play the result from start through success, failure, and restart. Check camera, input, objective clarity, feedback, interaction state, visual hierarchy, and obvious performance or stability problems.
Review the project, source, Blueprints, assets, plugins, rights, configuration, performance, saves, networking, cook, package, and target requirements before local continuation or external release.
Create a native Unreal 5 delivery-race prototype in a compact alpine village. The player collects one parcel, chooses between a safe road and a faster rooftop route, reaches the recipient before a visible timer expires, and can immediately retry. Keep the scope to one polished route decision, clear feedback, keyboard and controller support, and a browser-preview session under four minutes.
Snapshot date: July 25, 2026. Anthropic's July 24, 2026 announcement is the first-party source for release status, pricing, positioning, and launch availability. It does not claim an Epic-supported Unreal integration. Epic documentation and the exact project remain authoritative for engine behavior, compilation, assets, tests, cooking, packaging, and target-platform results.
Release date, claude-opus-5 model ID, coding and agent positioning, pricing, Fast mode, alignment, safety, and availability.
Open official announcementRe-check current model behavior, API surface, prompting guidance, limits, effort settings, and migration notes before production use.
Use version-specific Epic documentation and the real project for C++, Blueprint, build, test, cook, and package decisions.
Compare Claude Opus 5 and Opus 4.8 for Unreal coding, debugging, long agents, price, migration, safety fallback, native validation, and rollback.
Read this guideBuild focused Claude Opus 5 prompts for Unreal C++, Blueprints, logs, assets, and release tasks, then turn the approved brief into a native Unreal 5 project with SEELE AI.
Read this guideA complete Claude Opus 5 Unreal Engine guide covering the July 2026 release, coding and agent claims, pricing, limits, evaluation, and a direct SEELE game-creation path.
Read this guideAt launch, Anthropic lists $5 per million input tokens and $25 per million output tokens for the API.
The announced base token prices are the same. Actual task cost can differ because performance, effort, tokens, tools, retries, and review differ.
Anthropic says Fast mode is available at twice the base price and runs around 2.5 times the default speed. Verify current terms before budgeting.
Benchmark the lowest setting that reliably passes the task. Increase effort only when accepted-result or review gains justify the added tokens and latency.
Use Opus for approved reasoning or coding assistance; use SEELE when the desired output is a generated native Unreal project with browser preview and download.
Run repeated representative tasks and include tokens, tools, latency, builds, retries, review, test infrastructure, failures, and accepted-output rate; do not extrapolate from list price alone.
Open the canonical SEELE Unreal creator, choose a verified starter world, generate the native project, inspect the browser preview, then download or continue optimization and packaging. Keep third-party model evaluation and project evidence separate.