Free tools/LLM Pricing Comparison 2026

Compare LLM API prices

USD per million tokens · sortable table

Compare input rates, output rates, and context limits from a dated model-pricing snapshot.

GPT-6 LunaChecked 2026-11-12. Standard global API text rates. Above 272,000 input tokens, full-request input is 2× and output 1.5×. Cached input, tools, Fast/Batch and regional pricing excluded.lowest listed inputOpenAI$0.1$0.51,050k (~1575 pp)
Claude Haiku 5.5Checked 2026-11-12. Direct API base text rates. Caching, tools, Batch and regional/cloud pricing excluded. Above 100,000 input tokens: $0.50 input and $2.50 output per million.lowest listed inputAnthropic$0.1$0.51,000k (~1500 pp)
Gemini 3.8 FlashChecked 2026-11-12. Paid Standard text rates through 31 December 2026. From 1 January 2027: $1.50 input / $7.50 output per million. Output includes thinking; tools and caching excluded.Google$0.75$3.751,048.576k (~1573 pp)
GPT-6.1 SolChecked 2026-11-12. Standard global API text rates. Above 272,000 input tokens, full-request input is 2× and output 1.5×. Cached input, tools, Fast/Batch and regional pricing excluded.OpenAI$2$101,050k (~1575 pp)
Claude Sonnet 5.5Checked 2026-11-12. Direct API base text rates. Caching, tools, Batch and regional/cloud pricing excluded.Anthropic$2$101,000k (~1500 pp)
Claude Opus 5.5Checked 2026-11-12. Direct API base text rates. Caching, tools, Batch and regional/cloud pricing excluded.Anthropic$4$201,000k (~1500 pp)

Snapshot checked 2026-11-12. Selected models; not an exhaustive ranking. USD per 1M uncached text tokens, Standard global API processing. Base-tier prices appear above; the cost calculator applies documented long-prompt tiers. Follow each model link for its source. Click a column heading to sort.

How to use it

  1. Compare the relevant rate

    Sort by input price for large prompts or output price for long replies. Both contribute to the cost of a request.

  2. Check the model and limits

    Confirm the exact model identifier and whether its input and output limits suit your task.

  3. Estimate your workload

    Use the LLM API cost calculator with your expected tokens and monthly calls, then check current provider terms.

More questions

How current is the table?

The tool displays the snapshot date and data source. Prices can change between updates; this is not a live quote.

What can change the effective price?

Caching, batch processing, long-context pricing, tool calls, and provider-specific fees can affect the final bill.

Method and example

Compare standard uncached text-token rates for a selected set of models verified against provider documentation. Each row links to its source and states pricing conditions.

Illustrative example: One million input tokens at $2 plus one million output tokens at $10 cost $12 before other fees, provided the workload stays within that rate tier.

This is a selected comparison, not an exhaustive market ranking. Availability and prices can change; the lowest token price may not produce the lowest cost per accepted result.

More free AI tools.

Discuss your result with us

Bring the result and one workflow to a free 30-minute call. We will review the assumptions and discuss a practical next step.