OpenAI API vs Anthropic Claude API pricing
Both are in llm apis. OpenAI API is billed per token; Anthropic Claude API is billed per token. Numbers below are copied from each vendor's pricing page.
OpenAI APINo free tier
- Free tier
- $5 trial credit (after ID verification); opt-in data-sharing free tokens
- Billed
- Per token
- Best for
- Frontier general-purpose LLM platform
- Verified
- Jun 27, 2026
Anthropic Claude APINo free tier
- Free tier
- Trial credits on signup (no ongoing API free tier)
- Billed
- Per token
- Best for
- Frontier reasoning and agentic coding
- Verified
- Jun 27, 2026
OpenAI API plans
| Plan | Price | What it covers |
|---|---|---|
| GPT-5.5 (flagship) | $5.00 / $30.00 per 1M tokens | Input / output; cached input $0.50/1M. Highest-intelligence model (AA Index 55). |
| GPT-5.4 | $2.50 / $15.00 per 1M tokens | Input / output; cached input $0.25/1M. Mid-tier flagship. |
| GPT-5.4-mini | $0.75 / $4.50 per 1M tokens | Input / output; cached input $0.075/1M. Cost-efficient general model. |
| GPT-5.4-nano | $0.20 / $1.25 per 1M tokens | Input / output; cached input $0.02/1M. Cheapest, lowest-latency tier (~0.56s TTFT). |
| Batch / Flex processing | Up to ~50% off standard | Asynchronous Batch API and Flex tier trade latency for substantially lower per-token cost. |
| Scale Tier | Custom / committed throughput | Prioritized compute with a contractual 99.9% uptime SLA for production traffic. |
- Expensive output tokens on flagship models ($30/1M on GPT-5.5) make high-volume generation costly
- Standard pay-as-you-go tier has no uptime SLA; guarantees require Scale Tier or Azure
- Recurring billing-transparency and credit-reconciliation complaints from users
Anthropic Claude API plans
| Plan | Price | What it covers |
|---|---|---|
| Claude Opus 4.8 | $5 / $25 per 1M tokens | Flagship model, input/output; 1M-token context with no long-context surcharge. Fast Mode available at $10/$50. |
| Claude Sonnet 4.6 | $3 / $15 per 1M tokens | Best speed/intelligence balance, input/output; 1M-token context. |
| Claude Haiku 4.5 | $1 / $5 per 1M tokens | Fastest, most cost-effective tier, input/output; 200K context. |
| Claude Fable 5 | $10 / $50 per 1M tokens | Most capable widely released model for the most demanding reasoning/agentic work; 1M context. |
| Batch API | 50% off standard rates | Asynchronous processing; up to 100K requests or 256MB per batch, most complete within 1 hour. |
| Prompt caching | ~0.1x read / 1.25x write (5m) | Cached input served at ~10% of base price; up to ~90% savings on repeated prefixes. |
- Rate-limit and quota tiers are opaque, limits are not always clearly stated, generating developer frustration
- Premium pricing at the Opus/Fable tier is high versus commodity LLM APIs
- Some features are gated by model tier or unavailable on certain third-party platforms (e.g. no Batches/web search on Bedrock)
Source: https://platform.claude.com/docs/en/about-claude/pricing