Anthropic API review

The Anthropic API ranks third, close behind Gemini, with the clearest pricing page on the list: every Claude model, cache write and cache hit price sits in one USD table. Claude Fable 5.1 targets demanding reasoning and long-horizon agent work at $10 input and $50 output per 1M tokens. Opus 5.5 is $4 and $20 and is described for long-running agentic coding, Sonnet 5.5 is $2 and $10, and Haiku 5.5 is $0.10 input for prompts up to 100,000 tokens, rising to $0.50 above that. Prompt caching has 5-minute and 1-hour write options, and cache hits are a small fraction of input cost, which matters for agents that resend long context. The models are also sold through Amazon Bedrock and Google Cloud, so teams can buy through an existing cloud contract. Watch the tokenizer note: newer models produce about 30% more tokens for the same text, which raises the effective price.

Who it is for

Teams building coding tools, agents and long-document work who want Claude models with detailed caching controls.

Price

Pay per token; Claude Haiku 5.5 from $0.10 per 1M input tokens.

Claude Haiku 5.5 For high-volume, latency-sensitive tasks such as classification, extraction, and routing $0.10 / MTok for prompts up to 100,000 tokens
Quoted from the vendor pricing page on November 7, 2026.

Key features

• Four current models: Fable 5.1, Opus 5.5, Sonnet 5.5 and Haiku 5.5
• Prompt caching with 5-minute and 1-hour writes and low-cost cache hits
• Context windows up to 1M tokens, varying by model
• Available on the first-party API and through Amazon Bedrock and Google Cloud
• All prices published in USD on one docs page

Strengths and weaknesses

• Strength: Opus 5.5 at $4 input and $20 output per 1M tokens is aimed at long-running agentic coding
• Strength: Haiku 5.5 costs $0.10 per 1M input tokens for prompts up to 100,000 tokens
• Strength: Cache hits on Sonnet 5.5 cost $0.10 per 1M tokens
• Weakness: No permanent free API tier is listed
• Weakness: Haiku 5.5 input rises to $0.50 per 1M tokens for prompts over 100,000 tokens
• Weakness: The newer tokenizer produces about 30% more tokens for the same text, per the docs

Key facts

• Cheapest model: Claude Haiku 5.5, $0.10 input per 1M tokens up to 100,000-token prompts
• Mid tier: Claude Sonnet 5.5, $2 input and $10 output per 1M tokens
• Top model: Claude Fable 5.1, $10 input and $50 output per 1M tokens

Where it ranks

Anthropic API is No. 2 in our Best LLM APIs verdict.