Skip to content
Best AI Ever

Tool

Compare AI models

Pick two to four models. Every value comes from our structured data — prices, specs, capabilities, benchmarks and rankings — never typed twice.

Which rows matter depends on the job: how to compare AI models explains how to read benchmark scores with their setting and source and how to normalize prices for your token mix, and AI token pricing explained shows what the price rows mean for a real workload.

Claude Sonnet 5.5 vs GPT-6.1 Sol

Attribute Claude Sonnet 5.5Anthropic GPT-6.1 SolOpenAI
Pricing
Input tokensbase tier $2.00 per 1M tokens $2.00 per 1M tokens tiered
Cached input tokensbase tier $0.10 per 1M tokens $0.10 per 1M tokens tiered
Cache writesbase tier $2.50 per 1M tokens $2.50 per 1M tokens tiered
Cache writes (1-hour retention)base tier $4.00 per 1M tokens No data
Output tokensbase tier $10.00 per 1M tokens $10.00 per 1M tokens tiered
Batch input tokensbase tier $1.00 per 1M tokens $1.00 per 1M tokens tiered
Batch output tokensbase tier $5.00 per 1M tokens $5.00 per 1M tokens tiered
Tool and search requestseach price with what it covers
  • Web search tool on the Claude API: per 1,000 searches, plus standard token costs for search-generated content; a search that ends in an error is not billed: $10.00 per 1,000 requests
  • Web search tool (pricing page, Tools table: all models): per 1,000 calls, plus search content tokens billed at model rates: $10.00 per 1,000 requests
Othereach price with what it covers No data
  • 12-hour cache writes: $5.00 per 1M units 12-hour cache write (prompt_cache_options.ttl "12h", Responses API, added 2026-10-20), billed at 2.5x the uncached input rate; Standard, prompts up to 272K input tokens (pricing page)
  • 12-hour cache writes, > 272K input tokens: $10.00 per 1M units 12-hour cache write, long-context column of the pricing page Standard table (prompts with more than 272K input tokens)
Specs
Context window 1,000,000 tokens 1,050,000 tokens
Max output 128,000 tokens 128,000 tokens
License Proprietary Proprietary
Release date 2026-09-28 2026-09-29
Availability Available Available
Knowledge cutoff June 2026 April 2026
Capabilities
Text input Yes Yes
Image input Yes Yes
Audio input No No
Video input No No
File input Yes Yes
Text output Yes Yes
Image output No No
Audio output No No
Reasoning Optional Native
Tool calling Yes Yes
Web search Yes Yes
Best for
Codingrank in "Best AI for coding" #4 of 20 #3 of 20
Writing No data No data
Researchrank in "Best AI for research" #4 of 14 No data
Reasoningfrom structured data (no ranking mapped)
  • Reasoning: Optional
  • Reasoning: Native
Strengths and weaknesses
Strengths
  • 1M-token context window; 128K max output tokens (up to 300K on the Message Batches API with the output-300k beta header)
  • Comparative latency rated "Fast" at $2 / $10 per million input / output tokens
  • Cache reads at 0.05x base input ($0.10 per million tokens; reduced from $0.20 on 2026-10-07 per the API release notes)
  • Up-front thinking can be turned off with the between_tools setting at high effort or below
  • 1,050,000-token context window (maximum 922,000 input tokens) and 128,000 max output tokens
  • $2 input / $10 output per 1M tokens (Standard, up to 272K input tokens); cached input priced at 5% of the uncached input rate
  • Ultrafast mode available with US and EU data residency
  • Multi-agent (delegating work to subagents in a Responses API request) supported in beta
Weaknesses
  • Thinking type 'disabled' is rejected; between_tools is the lowest thinking setting
  • Forced tool use (tool_choice any/tool) and non-default temperature/top_p/top_k return a 400 error
  • Thinking blocks work only in the account that produced them or a linked account
  • No none or minimal reasoning effort
  • Tool calling requires the Responses API; Chat Completions is supported without tool calling
  • Audio and video modalities, fine-tuning and predicted outputs are not supported

"No data" means we have no verified value; it is never an estimate. Benchmarks are compared only within the same setting.