OpenAI · text · family comparison
GPT API pricing by model and provider
Compare GPT API routes by exact model ID, context band, input, cached-input and output meters, provider channel, and source.
Model-level provider comparison
GPT API price ranking status
Current answer: No current cheapest provider is published. Accepted evidence exists, but it does not yet form a named, like-for-like comparison group.
- Provider routes with evidence
- 3
- Accepted observations
- 8
- Billing meters
- 1M cached input tokens, 1M input tokens, 1M output tokens
- Latest accepted check
- Aug 28, 2026
Why there is no winner
No reviewed comparison scope exists for this model yet. Accepted evidence can still be inspected below, but APIDir will not sort unlike offers or infer a cheapest provider.
Read the publication gateUncollapsed source records
All accepted offer evidence
These rows remain separate until their offered model ID, route, meter, tier, resolution, duration, audio, and region are proven comparable.
| Provider | Model | Configuration | Recorded price | Evidence |
|---|---|---|---|---|
| OpenAI APIofficial | GPT API | model: gpt-5.6-luna · responses · standard-under-270k-context · not-applicable · no audio | $0.02 / 1M cached input tokensstale | Provider sourceChecked Aug 22, 2026 |
| OpenAI APIofficial | GPT API | model: gpt-5.6-luna · responses · standard-under-270k-context · not-applicable · no audio | $0.2 / 1M input tokensstale | Provider sourceChecked Aug 22, 2026 |
| OpenAI APIofficial | GPT API | model: gpt-5.6-luna · responses · standard-under-270k-context · not-applicable · no audio | $1.2 / 1M output tokensstale | Provider sourceChecked Aug 22, 2026 |
| OpenRouterrouter | GPT API | model: openai/gpt-5.6-luna · chat-completions · standard-under-270k-context · not-applicable · no audio | $0.2 / 1M input tokensstale | Provider sourceChecked Aug 22, 2026 |
| OpenRouterrouter | GPT API | model: openai/gpt-5.6-luna · chat-completions · standard-under-270k-context · not-applicable · no audio | $1.2 / 1M output tokensstale | Provider sourceChecked Aug 22, 2026 |
| XiuRouterrouter | GPT-5.4 | model: gpt-5.4 · listed-model-access · max service tier · not-applicable · no audio · source unit: USD per 1M tokens; XiuRouter max service tier; service capacity provided by XiuStore; protocol compatibility not inferred | $0.02 / 1M cached input tokensstale | Provider sourceChecked Aug 28, 2026 |
| XiuRouterrouter | GPT-5.4 | model: gpt-5.4 · listed-model-access · max service tier · not-applicable · no audio · source unit: USD per 1M tokens; XiuRouter max service tier; service capacity provided by XiuStore; protocol compatibility not inferred | $0.17 / 1M input tokensstale | Provider sourceChecked Aug 28, 2026 |
| XiuRouterrouter | GPT-5.4 | model: gpt-5.4 · listed-model-access · max service tier · not-applicable · no audio · source unit: USD per 1M tokens; XiuRouter max service tier; service capacity provided by XiuStore; protocol compatibility not inferred | $1.05 / 1M output tokensstale | Provider sourceChecked Aug 28, 2026 |
Checked evidence, not a ranking: fresh rows were checked against the linked first-party source on the shown date. APIDir does not name a cheapest provider unless exact configurations and units match. Recheck before purchasing.
Developer resources
Provider documentation and pricing links
Compare integration documentation beside the exact first-party source used for each accepted observation.
| Provider | Channel | Developer resources | Accepted evidence |
|---|---|---|---|
| OpenAI API | official | DocumentationPricing page | 3 observationsEvidence source 1 |
| OpenRouter | router | DocumentationPricing page | 2 observationsEvidence source 1 |
| XiuRouter | router | DocumentationPricing page | 3 observationsEvidence source 1 |
Decision guide
How to compare GPT API providers
GPT API pricing is a family-level search, but a production bill belongs to an exact model and service route. APIDir records GPT-5.6 Luna across direct OpenAI and OpenRouter channels, plus a separate XiuRouter gpt-5.4 tier. These different model IDs are evidence rows, not a like-for-like ranking, and new GPT versions must not silently replace them.
Input, cached input, and output use separate meters, and context thresholds can change rates. A workload with long prompts and short answers behaves differently from an agent that produces lengthy tool traces. Compare a weighted workload estimate, not one attractive token price.
What changes GPT API cost
- Exact GPT model, context band, input-output ratio, and cache hit rate
- Standard, batch, priority, or router service channel
- Tool calls, image or audio inputs, retries, rate limits, and retention requirements
Integration checks before production
- Keep the model ID and pricing tier in deploy configuration, not scattered literals
- Fail fast when the configured model or API credential is missing
- Log token usage, cache usage, route, retry count, and provider request ID for reconciliation
Workloads this comparison can inform
High-volume classification
Input volume dominates, so cached-input eligibility and batching can matter more than output price.
Coding and agent workloads
Budget for long outputs, tool loops, and context growth across turns.
Multi-provider resilience
A router can simplify failover, but its routing policy and service terms remain part of the product.
How to read the price result
Direct and router rows currently match the short-context token meters, but two channels are not enough for APIDir's three-provider winner gate. Long-context and non-text meters remain separate.
Related model price comparisons
Use this evidence safely
A recorded price is not a complete production quote. Open the source, reproduce the exact configuration, and include retries, failed work, storage, egress, support, rate limits, and downstream processing in total cost.
Questions developers ask
What makes two GPT offers comparable?
The exact model and endpoint, provider route, input and output meters, cache policy, context-length price band, tools, region, and service tier must match.
Does APIDir name a cheapest provider on this page?
Only when at least three explicitly reviewed provider offers are fresh, publication-permitted, and within the same published comparison scope. Stale or mismatched evidence cannot support that claim.
How should I confirm a budget?
Open the linked provider source, verify the configuration, then calculate your own successful and failed workload volume.
APIDir Delta · controlled rollout
Join the source-linked pricing update waitlist.
Confirm your address to join the waitlist. Editorial delivery starts only after sender, suppression, and evidence checks are active.