OpenAI API
Industry-standard foundation models with automatic prompt caching and vision.
Supported plans
Official pricingBase $0 / month equivalent. Metered charges and allowances are included at the default workload.
Conditions and limitations
- Flagship multimodal intelligence. Automatic 50% discount on cached prompt prefixes.
- Fine-tuning, Batch API discounts, audio/realtime APIs, and image generation.
Pricing checked 2026-10-06. Inspect source
Add this plan to a stackBase $0 / month equivalent. Metered charges and allowances are included at the default workload.
Conditions and limitations
- Automatic 50% prompt caching discount applies to prompts with 1,024+ identical leading tokens.
- Fine-tuning, Batch API discounts, audio/realtime APIs, and image generation.
Pricing checked 2026-10-06. Inspect source
Add this plan to a stackBase $0 / month equivalent. Metered charges and allowances are included at the default workload.
Conditions and limitations
- Reasoning tokens generated during thinking are billed as output completion tokens.
- Vision/multimodal inputs, streaming reasoning content without effort configuration.
Pricing checked 2026-10-06. Inspect source
Add this plan to a stackPricing observations
These dates mark when a price model was checked. They do not establish when a vendor changed its price. One observation is not a historical trend.
GPT-4o mini · Base $0 / month equivalent · Monthly billing
Official source ↗ · Usage charges are additional where modeled.GPT-4o mini · Base $0 / month equivalent · Monthly billing
Official source ↗ · Usage charges are additional where modeled.GPT-4o · Base $0 / month equivalent · Monthly billing
Official source ↗ · Usage charges are additional where modeled.GPT-4o · Base $0 / month equivalent · Monthly billing
Official source ↗ · Usage charges are additional where modeled.o3-mini · Base $0 / month equivalent · Monthly billing
Official source ↗ · Usage charges are additional where modeled.o3-mini · Base $0 / month equivalent · Monthly billing
Official source ↗ · Usage charges are additional where modeled.
Compare OpenAI API with Competitors
Leading coding and analytical models with 90% prompt cache read discounts.
Compare head-to-head ↗Ultra cost-effective open-weights models with deep reasoning and context caching.
Compare head-to-head ↗LPU Inference Engine delivering unmatched token output speeds.
Compare head-to-head ↗