Anthropic
Claude 5 — Fable 5.1, Opus 5, Sonnet 5, Haiku 4.5 with 1M context
Anthropic builds the Claude model family, now in its fifth generation. The Claude 5 lineup includes Fable 5.1 and Opus 5 at the frontier (1M+ token context, reasoning models), Claude Sonnet 5 for production workloads, and Haiku 4.5 for cost-efficient inference. Anthropic also offers Claude Code (agentic coding) and Claude Science (research) as specialised products. Prompt caching is available at up to 90% discount.
Claude 5 Opus and Fable 5 lead the BenchLM leaderboard as of September 2026, with 1M+ context windows and extended reasoning — plus Claude Code for agentic software engineering.
- Claude 5 family leads on coding and reasoning benchmarks
- 1M+ token context window on Claude 5 models
- Aggressive prompt caching — up to 90% discount
- Claude Code for agentic software engineering
- Extended thinking / reasoning mode on Opus 5 and Fable 5
- No open-weight models — full vendor lock-in
- Frontier models are among the higher-priced options — check live table for current rates
- No native image generation capability
Anthropic Models
| Model | Modality | Tier | Price | Cached /1M | TTFT | Throughput | Context | Intel. | Caps | |
|---|---|---|---|---|---|---|---|---|---|---|
claude-haiku-5-5 Claude 5 | Chat | efficient | $0.10/ $0.50 in / out per 1M EST Oct 2026 | $0.01 | 200ms | 250 t/s | 1.0M | 72 | ||
claude-haiku-3-20240307 Claude 3 | Chat | budget | $0.25/ $1.25 in / out per 1M EST Jul 2026 | $0.03 | 250ms | 280 t/s | 200K | 52 | ||
claude-haiku-3 Claude 3 | Chat | budget | $0.25/ $1.25 in / out per 1M EST Jul 2026 | $0.03 | 200ms | 250 t/s | 200K | 60 | ||
claude-3-haiku-20240307 Claude 3 | Chat | budget | $0.25/ $1.25 in / out per 1M EST Jul 2026 | $0.03 | 200ms | 180 t/s | 200K | 65 | ||
claude-haiku-3-5 Claude 3.5 | Chat | efficient | $0.80/ $4.00 in / out per 1M EST Jul 2026 | $0.08 | 320ms | 200 t/s | 200K | 65 | ||
claude-haiku-4.5 Claude 4 | Chat | efficient | $1.00/ $5.00 in / out per 1M EST Jul 2026 | $0.10 | 250ms | 220 t/s | 200K | 74 | ||
claude-sonnet-5 Claude 5 | Chat | frontier | $2.00/ $10.00 in / out per 1M EST Jul 2026 | $0.20 | 700ms | 95 t/s | 1.0M | 90 | ||
claude-sonnet-5-5 Claude 5 | Chat | frontier | $2.00/ $10.00 in / out per 1M EST Oct 2026 | $0.10 | 700ms | 95 t/s | 1.0M | 91 | ||
claude-sonnet-4-5 Claude 4 | Chat | advanced | $3.00/ $15.00 in / out per 1M EST Jul 2026 | $0.30 | 650ms | 120 t/s | 200K | 82 | ||
claude-3.5-sonnet-20241022 Claude 3.5 | Chat | frontier | $3.00/ $15.00 in / out per 1M EST Jul 2026 | $0.30 | 600ms | 120 t/s | 200K | 82 | ||
claude-3.7-sonnet-20250219 Claude 3.7 | Chat | frontier | $3.00/ $15.00 in / out per 1M EST Jul 2026 | $0.30 | 700ms | 110 t/s | 200K | 85 | ||
claude-sonnet-4-20250514 Claude 4 | Chat | advanced | $3.00/ $15.00 in / out per 1M EST Jul 2026 | $0.30 | 620ms | 125 t/s | 200K | 84 | ||
claude-sonnet-4-6 Claude 4 | Chat | frontier | $3.00/ $15.00 in / out per 1M EST Jul 2026 | $0.30 | 600ms | 100 t/s | 200K | 88 | ||
claude-opus-5-5 Claude 5 | Chat | frontier | $4.00/ $20.00 in / out per 1M EST Oct 2026 | $0.20 | 1.2s | 53 t/s | 1.0M | 97 | ||
claude-opus-4-6 Claude 4 | Chat | frontier | $5.00/ $25.00 in / out per 1M EST Jul 2026 | $0.50 | 900ms | 70 t/s | 200K | 91 | ||
claude-opus-4-8 Claude 4 | Chat | frontier | $5.00/ $25.00 in / out per 1M EST Jul 2026 | $0.50 | 1.0s | 65 t/s | 1.0M | 92 | ||
claude-opus-5 Claude 5 | Chat | frontier | $5.00/ $25.00 in / out per 1M EST Aug 2026 | $0.50 | 1.2s | 53 t/s | 1.0M | 97 | ||
claude-mythos-5 Claude 5 | Chat | frontier | $10.00/ $50.00 in / out per 1M EST Aug 2026 | $1.00 | 2.0s | 45 t/s | 1.0M | 98 | ||
claude-fable-5 Claude 5 | Chat | frontier | $10.00/ $50.00 in / out per 1M EST Aug 2026 | $1.00 | 1.8s | 68 t/s | 1.0M | 96 | ||
claude-fable-5-1 Claude 5 | Chat | frontier | $10.00/ $50.00 in / out per 1M EST Oct 2026 | $0.25 | 1.8s | 68 t/s | 1.0M | 96 | ||
claude-mythos-5-1 Claude 5 | Chat | frontier | $10.00/ $50.00 in / out per 1M EST Oct 2026 | $0.25 | 2.0s | 45 t/s | 1.0M | 98 | ||
claude-opus-4-5 Claude 4 | Chat | frontier | $15.00/ $75.00 in / out per 1M EST Jul 2026 | $1.50 | 1.2s | 70 t/s | 200K | 90 | ||
claude-opus-4-20250514 Claude 4 | Chat | frontier | $15.00/ $75.00 in / out per 1M EST Jul 2026 | $1.50 | 1.1s | 75 t/s | 200K | 91 | ||
claude-opus-3 Claude 3 | Chat | frontier | $15.00/ $75.00 in / out per 1M EST Jul 2026 | $1.50 | 1.2s | 45 t/s | 200K | 80 | ||
claude-3-opus-20240229 Claude 3 | Chat | frontier | $15.00/ $75.00 in / out per 1M EST Jul 2026 | $1.50 | 800ms | 30 t/s | 200K | 82 |
Anthropic — Frequently Asked Questions (FAQs)
Similar Providers
Anthropic Reviews
All LLM Providers
Full comparison tableAnthropic LLM pricing overview
Anthropic is a proprietary AI lab that develops and hosts its own large language models. Anthropic builds the Claude model family, now in its fifth generation. The Claude 5 lineup includes Fable 5.1 and Opus 5 at the frontier (1M+ token context, reasoning models), Claude Sonnet 5 for production workloads, and Haiku 4.5 for cost-efficient inference. Anthropic also offers Claude Code (agentic coding) and Claude Science (research) as specialised products. Prompt caching is available at up to 90% discount. Common use cases include Coding, Reasoning, Long-context, Chat. Headquartered in San Francisco, CA, founded 2021. All models are accessible via a REST API compatible with standard OpenAI-style request formats, enabling drop-in integration with most LLM frameworks and orchestration tools.
Anthropic vs other LLM providers
Anthropic competes with OpenAI, Anthropic, Google, Groq, Together AI, Mistral, Cohere, and other inference API providers across dimensions of price, throughput, latency, context window, and model intelligence. Claude 5 Opus and Fable 5 lead the BenchLM leaderboard as of September 2026, with 1M+ context windows and extended reasoning — plus Claude Code for agentic software engineering. Use the LLM provider comparison tool to see Anthropic token pricing, latency, and throughput side-by-side with any other provider. The full LLM pricing table shows all providers ranked by input token cost, output token cost, and throughput in a single sortable view.
Why choose Anthropic?
Anthropic stands out because: Claude 5 Opus and Fable 5 lead the BenchLM leaderboard as of September 2026, with 1M+ context windows and extended reasoning — plus Claude Code for agentic software engineering.. The API currently offers 25 models, spanning 17 frontier, 2 advanced, 3 efficient, 3 budget tiers. For teams running high-volume inference workloads, prompt caching and batch API endpoints can reduce effective input token costs by 50–90% — check the context window cost guide for a full breakdown of caching economics.
Understanding Anthropic token pricing
Anthropic charges separately for input (prompt) and output (completion) tokens, priced per 1M tokens in USD. Output tokens are typically 3–5× more expensive than input tokens due to the compute cost of autoregressive generation. The cheapest {meta.name} input rate is currently $0.10/1M tokens. Prices shown are sourced from Anthropic's public pricing page and updated every 15 minutes. Need help estimating your spend? Read the LLM API cost calculator guide — it covers tokens, context windows, prompt caching, and batch discounts with worked examples for RAG, chat history, and document processing workloads.
Anthropic context window and model capabilities
Context window size directly affects both capability and cost — every token in the context window is charged as an input token. Anthropic's largest context window is 1.0M tokens. For RAG and document processing workloads, longer context windows enable richer retrieval but increase per-call costs proportionally. Prompt caching — where supported — stores the KV state of repeated prefixes and charges 75–90% less for cache hits, making it the most impactful cost optimization for applications with consistent system prompts or retrieved documents. See the LLM context window cost guide for a full analysis of how context length affects your API bill.
Self-hosted vs managed inference: when Anthropic makes sense
Managed inference APIs like Anthropic eliminate infrastructure overhead — no GPU provisioning, driver management, or model serving stack to maintain. The trade-off is cost at scale: a single H100 at ~$2.50/hr can serve ~500K tokens/min of Llama 3.3 70B, which at Anthropic API rates would cost significantly more per token. The break-even point depends on your request volume, latency requirements, and engineering capacity. For teams processing fewer than ~10M tokens/day, managed APIs are almost always cheaper when total cost of ownership is considered. Above that threshold, self-hosted inference on rented GPU compute typically wins on unit economics. Read the cheapest GPU cloud guide for a full break-even analysis. Historical Anthropic token price data is available in the LLM price history charts.
Compare Anthropic With Another LLM Provider
Compare Anthropic vs other providers
Side-by-side token pricing, latency, throughput, and context window. View all LLM provider comparisons →
Looking for alternatives?
Compare Anthropic with other LLM API providers side by side.