AI Pricing Intelligence

Compare AI API prices, cloud model routes, and AI product subscriptions in one place. PlanPulse separates monthly plan price from real usage mechanics, caveats, and source confidence.

Showing 14 rows · 0 selected for quick comparison

Compare	Provider / Channel	Model	Input	Cached	Output	Context	Best For	Confidence
Select	OpenAIDirect API	gpt-5.5text + image input, text output	$5.00 short / $10.00 long	$0.50 short / $1.00 long	$30.00 short / $45.00 long	1MMax output: 128K	Frontier coding, reasoning, professional agents	High Source
Select	OpenAIDirect API	gpt-5.4-minitext + image input, text output	$0.75	$0.075	$4.50	400KMax output: 128K	Lower-latency coding, computer use, subagents	High Source
Select	AnthropicClaude API	Claude Opus 4.8text + vision/PDF input, text output	$5.00	$0.50 cache read	$25.00	1MMax output: 128K for 1M-context models	Hard reasoning, complex agents, coding	High Source
Select	AnthropicClaude API	Claude Sonnet 4.6text + vision/PDF input, text output	$3.00	$0.30 cache read	$15.00	1MMax output: 128K for 1M-context models	Balanced coding, agents, production workflows	High Source
Select	AnthropicClaude API	Claude Haiku 4.5text + vision/PDF input, text output	$1.00	$0.10 cache read	$5.00	200KMax output: Needs Review	Fast extraction, routing, simple agents	Medium Source
Select	Google GeminiGemini Developer API	gemini-2.5-protext/image/video/audio/PDF input, text output	$1.25 <=200K / $2.50 >200K	$0.125 <=200K / $0.25 >200K plus storage	$10.00 <=200K / $15.00 >200K	1,048,576 input tokensMax output: 65,536	Complex reasoning, coding, large document analysis	High Source
Select	Google GeminiGemini Developer API	gemini-2.5-flashtext/image/video/audio input, text output	$0.30 text/image/video; $1.00 audio	$0.03 text/image/video; $0.10 audio plus storage	$2.50	1,048,576 input tokensMax output: 65,536	Low-latency high-volume reasoning tasks	High Source
Select	DeepSeekDirect API	deepseek-v4-flashtext input/output	$0.14 cache miss	$0.0028 cache hit	$0.28	1MMax output: 384K	Very low-cost coding/chat/agent throughput	High Source
Select	DeepSeekDirect API	deepseek-v4-protext input/output	$0.435 cache miss	$0.003625 cache hit	$0.87	1MMax output: 384K	Higher-quality DeepSeek reasoning/coding at low cost	High Source
Select	MistralLa Plateforme API	Mistral Medium 3.5text-to-text, reasoning, coding, agentic, multimodal	$1.50	Not listed	$7.50	Needs ReviewMax output: Needs Review	Frontier-class agentic and coding use cases	Medium Source
Select	xAIDirect API	grok-4.3text input/output	$1.25	Not listed	$2.50	1MMax output: Needs Review	Flagship reasoning, tool calling, agentic workflows	High Source
Select	OpenRouterOpenRouter API	anthropic/claude-sonnet-4.6text/vision via model support	$3.00	Effective prompt caching may reduce cost; exact cached row not listed	$15.00	1MMax output: Needs Review	Claude Sonnet through router/fallback	Medium Source
Select	Azure OpenAIAzure OpenAI / Foundry	GPT-5.5 Globaltext + image input, text output	$5.00	$0.50	$30.00	Needs ReviewMax output: Needs Review	OpenAI model through Azure governance/networking	Needs Review Source
Select	AWS BedrockAmazon Bedrock	varies by provider/modelvaries	Needs Review	Needs Review	Needs Review	Needs ReviewMax output: Needs Review	Multi-provider enterprise AWS deployment	Needs Review Source

API purchasing table

Channel is a first-class field, so Claude through Anthropic, Bedrock, Vertex, Azure, or OpenRouter can be compared as commercially different supply paths.

Open table

Product plan table

Subscription rows track monthly cost, annual cost, included models, team features, API inclusion, entry points, and the hidden usage model behind each product.

Open plans

Source confidence

Each row links to its official source. Cloud-provider and fast-changing quota rows stay marked as Needs Review until model, region, and deployment mode are pinned down.

Read llms.txt