Pricing database
AI API pricing
Current prices per input, cached and output token and every other dimension we track, in the currency each provider publishes. Every price links its official source and shows when we last verified it.
Latest price check:
New to token prices? AI token pricing explained shows how input, cached and output tokens add up, and how to reduce AI API costs covers caching, batching and routing requests to cheaper models. To turn these prices into a monthly bill, use the cost calculator.
Input tokens
base tier, per 1M tokensCached input tokens
base tier, per 1M tokensCache writes
base tier, per 1M tokensCache writes (1-hour retention)
base tier, per 1M tokens| # | Model | Provider | Price | Verified | Source |
|---|---|---|---|---|---|
| 1 | Claude Haiku 5.5 | Anthropic | $0.20 per 1M tokens tiered | 2026-10-25 | Pricing - Claude Platform Docs |
| 2 | Claude Sonnet 5.5 | Anthropic | $4.00 per 1M tokens | 2026-10-25 | Pricing - Claude Platform Docs |
| 3 | Claude Sonnet 4.5 | Anthropic | $6.00 per 1M tokens | 2026-10-25 | Pricing - Claude Platform Docs |
| 4 | Claude Opus 5.5 | Anthropic | $8.00 per 1M tokens | 2026-10-25 | Pricing - Claude Platform Docs |
| 5 | Claude Fable 5.1 | Anthropic | $20.00 per 1M tokens | 2026-10-25 | Pricing - Claude Platform Docs |
Cache storage
base tier, per 1M tokens per hour| # | Model | Provider | Price | Verified | Source |
|---|---|---|---|---|---|
| 1 | Gemini 3.6 Flash | $0.50 per 1M tokens per hour | 2026-10-25 | Gemini Developer API pricing | |
| 2 | Gemini 3.8 Flash | $0.50 per 1M tokens per hour | 2026-10-25 | Gemini Developer API pricing | |
| 3 | Gemini 3.1 Flash-Lite | $1.00 per 1M tokens per hour | 2026-10-25 | Gemini Developer API pricing | |
| 4 | Gemini 3.5 Flash-Lite | $1.00 per 1M tokens per hour | 2026-10-25 | Gemini Developer API pricing | |
| 5 | Gemini 3.1 Pro | $4.50 per 1M tokens per hour | 2026-10-25 | Gemini Developer API pricing |
Output tokens
base tier, per 1M tokensBatch input tokens
base tier, per 1M tokensBatch output tokens
base tier, per 1M tokensImage input tokens
base tier, per 1M tokens| # | Model | Provider | Price | Verified | Source |
|---|---|---|---|---|---|
| 1 | GPT-Image-2.5 Flare | OpenAI | $8.00 per 1M tokens | 2026-10-25 | Pricing | OpenAI API |
| 2 | GPT-Image-2.5 Sunburst | OpenAI | $8.00 per 1M tokens | 2026-10-25 | Pricing | OpenAI API |
Image output tokens
base tier, per 1M tokens| # | Model | Provider | Price | Verified | Source |
|---|---|---|---|---|---|
| 1 | GPT-Image-2.5 Flare | OpenAI | $30.00 per 1M tokens | 2026-10-25 | Pricing | OpenAI API |
| 2 | GPT-Image-2.5 Sunburst | OpenAI | $30.00 per 1M tokens | 2026-10-25 | Pricing | OpenAI API |
| 3 | Nano Banana 2 Lite | $30.00 per 1M tokens | 2026-10-25 | Gemini Developer API pricing | |
| 4 | Nano Banana 2.1 | $30.00 per 1M tokens | 2026-10-25 | Gemini Developer API pricing | |
| 5 | Nano Banana Pro | $120.00 per 1M tokens | 2026-10-25 | Gemini Developer API pricing |
Image generation
base tier, per image| # | Model | Provider | Price | Verified | Source |
|---|---|---|---|---|---|
| 1 | Muse Image | Meta | $0.01 per image | 2026-10-25 | Pricing and rate limits - Meta Model API |
| 2 | Grok Imagine Image | xAI | $0.02 per image | 2026-10-25 | API Pricing |
| 3 | Nano Banana 2 Lite | $0.0336 per image | 2026-10-25 | Gemini Developer API pricing | |
| 4 | Nano Banana 2.1 | $0.0336 per image | 2026-10-25 | Gemini Developer API pricing | |
| 5 | Grok Imagine Image 2.0 | xAI | $0.04 per image | 2026-10-25 | API Pricing |
| 6 | Qwen-Image-3.0-Pro | Alibaba Cloud (Qwen) | $0.04 per image | 2026-10-25 | qwen-image-3.0-pro Model Info - Alibaba Cloud Model Studio |
| 7 | Grok Imagine Image Quality | xAI | $0.05 per image | 2026-10-25 | API Pricing |
| 8 | Wan2.7-Image-Pro | Alibaba Cloud (Qwen) | $0.075 per image | 2026-10-25 | wan2.7-image-pro Model Info - Alibaba Cloud Model Studio |
| 9 | Nano Banana Pro | $0.134 per image | 2026-10-25 | Gemini Developer API pricing |
Audio input tokens
base tier, per 1M tokens| # | Model | Provider | Price | Verified | Source |
|---|---|---|---|---|---|
| 1 | Gemini 3.1 Flash-Lite | $0.50 per 1M tokens | 2026-10-25 | Gemini Developer API pricing | |
| 2 | Amazon Nova 2 Sonic | Amazon | $3.00 per 1M tokens | 2026-10-25 | Amazon Bedrock Pricing |
| 3 | Amazon Nova 2.5 Sonic | Amazon | $3.00 per 1M tokens | 2026-10-25 | Amazon Bedrock Pricing |
Audio output tokens
base tier, per 1M tokens| # | Model | Provider | Price | Verified | Source |
|---|---|---|---|---|---|
| 1 | Amazon Nova 2 Sonic | Amazon | $12.00 per 1M tokens | 2026-10-25 | Amazon Bedrock Pricing |
| 2 | Amazon Nova 2.5 Sonic | Amazon | $12.00 per 1M tokens | 2026-10-25 | Amazon Bedrock Pricing |
Tool and search requests
listed with what each price covers; not ranked| Model | Provider | Covers | Price | Verified | Source |
|---|---|---|---|---|---|
| Amazon Nova 2 Lite | Amazon | — US East (N. Virginia), on-demand list price; Web Grounding built-in tool, per 1,000 requests. | $30.00 per 1,000 requests | 2026-10-25 | Amazon Bedrock Pricing |
| Claude Fable 5.1 | Anthropic | — 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 | 2026-10-25 | Pricing - Claude Platform Docs |
| Claude Haiku 5.5 | Anthropic | — 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 | 2026-10-25 | Pricing - Claude Platform Docs |
| Claude Opus 5.5 | Anthropic | — 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 | 2026-10-25 | Pricing - Claude Platform Docs |
| Claude Sonnet 5.5 | Anthropic | — 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 | 2026-10-25 | Pricing - Claude Platform Docs |
| DeepSeek-V4-Pro | DeepSeek | Search / Fetch API Separate Search and Fetch API endpoints: per 1,000 successful requests each; failed, empty and timed-out requests are not billed. Calls by server-side tools in DeepSeek Harness, Claude Code or Codex are free until billing starts at 16:00 UTC on 2026-10-31 (00:00 Beijing time, 2026-11-01). | $0.75 per 1,000 requests | 2026-10-25 | Models & Pricing | DeepSeek API Docs |
| DeepSeek-V4.1-Flash | DeepSeek | Search / Fetch API Separate Search and Fetch API endpoints: per 1,000 successful requests each; failed, empty and timed-out requests are not billed. Calls by server-side tools in DeepSeek Harness, Claude Code or Codex are free until billing starts at 16:00 UTC on 2026-10-31 (00:00 Beijing time, 2026-11-01). | $0.75 per 1,000 requests | 2026-10-25 | Models & Pricing | DeepSeek API Docs |
| Gemini 3.1 Flash-Lite | Grounding with Google Maps Paid tier: 5,000 free requests per month (shared across all Gemini 3 and newer models), then USD 14 per 1,000 search queries | $14.00 per 1,000 requests | 2026-10-25 | Gemini Developer API pricing | |
| Gemini 3.1 Flash-Lite | Grounding with Google Search Paid tier: 5,000 free search requests per month (shared across all Gemini 3 and newer models), then this price; each Google Search query triggered by a request is charged | $14.00 per 1,000 requests | 2026-10-25 | Gemini Developer API pricing | |
| Gemini 3.1 Pro | Grounding with Google Maps Paid tier: 5,000 free requests per month (shared across all Gemini 3 and newer models), then USD 14 per 1,000 search queries | $14.00 per 1,000 requests | 2026-10-25 | Gemini Developer API pricing | |
| Gemini 3.1 Pro | Grounding with Google Search Paid tier: 5,000 free search requests per month (shared across all Gemini 3 and newer models), then this price; each Google Search query triggered by a request is charged | $14.00 per 1,000 requests | 2026-10-25 | Gemini Developer API pricing | |
| Gemini 3.5 Flash-Lite | Grounding with Google Maps Paid tier: 5,000 free requests per month (shared across all Gemini 3 and newer models), then USD 14 per 1,000 search queries | $14.00 per 1,000 requests | 2026-10-25 | Gemini Developer API pricing | |
| Gemini 3.5 Flash-Lite | Grounding with Google Search Paid tier: 5,000 free search requests per month (shared across all Gemini 3 and newer models), then this price; each Google Search query triggered by a request is charged | $14.00 per 1,000 requests | 2026-10-25 | Gemini Developer API pricing | |
| Gemini 3.6 Flash | Grounding with Google Maps Paid tier: 5,000 free requests per month (shared across all Gemini 3 and newer models), then USD 14 per 1,000 search queries | $14.00 per 1,000 requests | 2026-10-25 | Gemini Developer API pricing | |
| Gemini 3.6 Flash | Grounding with Google Search Paid tier: 5,000 free search requests per month (shared across all Gemini 3 and newer models), then this price; each Google Search query triggered by a request is charged | $14.00 per 1,000 requests | 2026-10-25 | Gemini Developer API pricing | |
| Gemini 3.8 Flash | Grounding with Google Maps Paid tier: 5,000 free requests per month (shared across all Gemini 3 and newer models), then USD 14 per 1,000 search queries | $14.00 per 1,000 requests | 2026-10-25 | Gemini Developer API pricing | |
| Gemini 3.8 Flash | Grounding with Google Search Paid tier: 5,000 free search requests per month (shared across all Gemini 3 and newer models), then this price; each Google Search query triggered by a request is charged | $14.00 per 1,000 requests | 2026-10-25 | Gemini Developer API pricing | |
| Gemini 4 Argon | Grounding with Google Maps Paid tier: 5,000 free requests per month (shared across all Gemini 3 and newer models), then USD 14 per 1,000 search queries | $14.00 per 1,000 requests | 2026-10-25 | Gemini Developer API pricing | |
| Gemini 4 Argon | Grounding with Google Search Paid tier: 5,000 free search requests per month (shared across all Gemini 3 and newer models), then this price; each Google Search query triggered by a request is charged | $14.00 per 1,000 requests | 2026-10-25 | Gemini Developer API pricing | |
| GPT-6 Astra | OpenAI | — 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 | 2026-10-25 | Pricing | OpenAI API |
| GPT-6 Luna | OpenAI | — 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 | 2026-10-25 | Pricing | OpenAI API |
| GPT-6 Sol | OpenAI | — 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 | 2026-10-25 | Pricing | OpenAI API |
| GPT-6.1 Sol | OpenAI | — 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 | 2026-10-25 | Pricing | OpenAI API |
| Muse Spark 1.1 | Meta | — Web search grounding (web_search tool): per 1,000 search queries, on top of token costs | $2.50 per 1,000 requests | 2026-10-25 | Pricing and rate limits - Meta Model API |
| Muse Spark 1.2 | Meta | — Web search grounding (web_search tool): per 1,000 search queries, on top of token costs | $2.50 per 1,000 requests | 2026-10-25 | Pricing and rate limits - Meta Model API |
| Muse Spark 1.3 | Meta | — Web search grounding (web_search tool): per 1,000 search queries, on top of token costs | $2.50 per 1,000 requests | 2026-10-25 | Pricing and rate limits - Meta Model API |
| Nano Banana 2.1 | — Grounding with Google Web and Image Search, paid tier: 5,000 free search requests per month (shared across all Gemini 3 and newer models), then this price; retrieved context is not charged as input tokens | $14.00 per 1,000 requests | 2026-10-25 | Gemini Developer API pricing | |
| Nano Banana Pro | — Grounding with Google Search, paid tier: 5,000 free search requests per month (shared across all Gemini 3 and newer models), then this price | $14.00 per 1,000 requests | 2026-10-25 | Gemini Developer API pricing | |
| Qwen3.7-Plus | Alibaba Cloud (Qwen) | — Singapore: 'agent' search strategy, or the web_search tool via the Responses API or Anthropic-compatible method; per 1,000 calls, plus retrieved content billed as input tokens. No fee is listed for the default turbo/max strategies | $10.00 per 1,000 requests | 2026-10-25 | Web search with large models - Alibaba Cloud Model Studio |
| Qwen3.8-2.4T-A95B | Alibaba Cloud (Qwen) | — Singapore: 'agent' search strategy, or the web_search tool via the Responses API or Anthropic-compatible method; per 1,000 calls, plus retrieved content billed as input tokens. No fee is listed for the default turbo/max strategies | $10.00 per 1,000 requests | 2026-10-25 | Web search with large models - Alibaba Cloud Model Studio |
| Qwen3.8-27B | Alibaba Cloud (Qwen) | — Singapore: 'agent' search strategy, or the web_search tool via the Responses API or Anthropic-compatible method; per 1,000 calls, plus retrieved content billed as input tokens. No fee is listed for the default turbo/max strategies | $10.00 per 1,000 requests | 2026-10-25 | Web search with large models - Alibaba Cloud Model Studio |
| Qwen3.8-Flash | Alibaba Cloud (Qwen) | — Singapore: 'agent' search strategy, or the web_search tool via the Responses API or Anthropic-compatible method; per 1,000 calls, plus retrieved content billed as input tokens. No fee is listed for the default turbo/max strategies | $10.00 per 1,000 requests | 2026-10-25 | Web search with large models - Alibaba Cloud Model Studio |
| Qwen3.8-Max | Alibaba Cloud (Qwen) | — Singapore: 'agent' search strategy, or the web_search tool via the Responses API or Anthropic-compatible method; per 1,000 calls, plus retrieved content billed as input tokens. No fee is listed for the default turbo/max strategies | $10.00 per 1,000 requests | 2026-10-25 | Web search with large models - Alibaba Cloud Model Studio |
| Qwen3.8-Omni-Flash | Alibaba Cloud (Qwen) | — Singapore: 'agent' search strategy, or the web_search tool via the Responses API or Anthropic-compatible method; per 1,000 calls, plus retrieved content billed as input tokens. No fee is listed for the default turbo/max strategies | $10.00 per 1,000 requests | 2026-10-25 | Web search with large models - Alibaba Cloud Model Studio |
Other
listed with what each price covers; not ranked| Model | Provider | Covers | Price | Verified | Source |
|---|---|---|---|---|---|
| Amazon Nova 2 Lite | Amazon | Flex input US East (N. Virginia), on-demand list price; Flex tier, Global cross-Region inference (Geo cross-Region/in-Region: 0.165 USD per 1M tokens). | $0.15 per 1M units | 2026-10-25 | Amazon Bedrock Pricing |
| Amazon Nova 2 Lite | Amazon | Flex output US East (N. Virginia), on-demand list price; Flex tier, Global cross-Region inference (Geo cross-Region/in-Region: 1.375 USD per 1M tokens). | $1.25 per 1M units | 2026-10-25 | Amazon Bedrock Pricing |
| Amazon Nova 2 Lite | Amazon | Priority input US East (N. Virginia), on-demand list price; Priority tier, Global cross-Region inference (Geo cross-Region/in-Region: 0.5775 USD per 1M tokens). | $0.525 per 1M units | 2026-10-25 | Amazon Bedrock Pricing |
| Amazon Nova 2 Lite | Amazon | Priority output US East (N. Virginia), on-demand list price; Priority tier, Global cross-Region inference (Geo cross-Region/in-Region: 4.8125 USD per 1M tokens). | $4.375 per 1M units | 2026-10-25 | Amazon Bedrock Pricing |
| Amazon Nova Pro | Amazon | Flex input US East (N. Virginia), on-demand list price; Flex tier, Geo cross-Region and in-Region inference. | $0.40 per 1M units | 2026-10-25 | Amazon Bedrock Pricing |
| Amazon Nova Pro | Amazon | Flex output US East (N. Virginia), on-demand list price; Flex tier, Geo cross-Region and in-Region inference. | $1.60 per 1M units | 2026-10-25 | Amazon Bedrock Pricing |
| Amazon Nova Pro | Amazon | Priority input US East (N. Virginia), on-demand list price; Priority tier, Geo cross-Region and in-Region inference. | $1.40 per 1M units | 2026-10-25 | Amazon Bedrock Pricing |
| Amazon Nova Pro | Amazon | Priority output US East (N. Virginia), on-demand list price; Priority tier, Geo cross-Region and in-Region inference. | $5.60 per 1M units | 2026-10-25 | Amazon Bedrock Pricing |
| Claude Opus 5.5 | Anthropic | Fast mode input Fast mode (out of research preview since 2026-10-15): Claude API, Claude Models on Google Cloud (Anthropic managed) and Microsoft Foundry deployments hosted on Azure; not available with the Batch API; applies across the full context window; prompt caching and data residency multipliers apply on top | $8.00 per 1M units | 2026-10-25 | Pricing - Claude Platform Docs |
| Claude Opus 5.5 | Anthropic | Fast mode output Fast mode (out of research preview since 2026-10-15): Claude API, Claude Models on Google Cloud (Anthropic managed) and Microsoft Foundry deployments hosted on Azure; not available with the Batch API; applies across the full context window; prompt caching and data residency multipliers apply on top | $40.00 per 1M units | 2026-10-25 | Pricing - Claude Platform Docs |
| DeepSeek-V4-Pro | DeepSeek | Off-peak cached input Cache hit. Off-peak rate (half of the peak rate): all hours outside 01:00-04:00 and 06:00-10:00 UTC Monday to Friday, including weekends and Chinese public holidays | $0.022 per 1M units | 2026-10-25 | Models & Pricing | DeepSeek API Docs |
| DeepSeek-V4-Pro | DeepSeek | Off-peak input Cache miss. Off-peak rate (half of the peak rate): all hours outside 01:00-04:00 and 06:00-10:00 UTC Monday to Friday, including weekends and Chinese public holidays | $0.66 per 1M units | 2026-10-25 | Models & Pricing | DeepSeek API Docs |
| DeepSeek-V4-Pro | DeepSeek | Off-peak output Off-peak rate (half of the peak rate): all hours outside 01:00-04:00 and 06:00-10:00 UTC Monday to Friday, including weekends and Chinese public holidays | $1.98 per 1M units | 2026-10-25 | Models & Pricing | DeepSeek API Docs |
| DeepSeek-V4.1-Flash | DeepSeek | Off-peak cached input Cache hit. Off-peak rate (half of the peak rate): all hours outside 01:00-04:00 and 06:00-10:00 UTC Monday to Friday, including weekends and Chinese public holidays | $0.003 per 1M units | 2026-10-25 | Models & Pricing | DeepSeek API Docs |
| DeepSeek-V4.1-Flash | DeepSeek | Off-peak input Cache miss. Off-peak rate (half of the peak rate): all hours outside 01:00-04:00 and 06:00-10:00 UTC Monday to Friday, including weekends and Chinese public holidays | $0.15 per 1M units | 2026-10-25 | Models & Pricing | DeepSeek API Docs |
| DeepSeek-V4.1-Flash | DeepSeek | Off-peak output Off-peak rate (half of the peak rate): all hours outside 01:00-04:00 and 06:00-10:00 UTC Monday to Friday, including weekends and Chinese public holidays | $0.60 per 1M units | 2026-10-25 | Models & Pricing | DeepSeek API Docs |
| Gemini 3.1 Flash-Lite | Batch audio input tokens Batch API price for audio input as listed on the pricing page (50% of Standard) | $0.25 per 1M units | 2026-10-25 | Gemini Developer API pricing | |
| Gemini 3.1 Flash-Lite | Cached audio input tokens Context caching price for audio input, Standard | $0.05 per 1M units | 2026-10-25 | Gemini Developer API pricing | |
| GPT-6 Astra | OpenAI | 12-hour cache writes 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) | $25.00 per 1M units | 2026-10-25 | Pricing | OpenAI API |
| GPT-6 Astra | OpenAI | 12-hour cache writes, > 272K input tokens 12-hour cache write, long-context column of the pricing page Standard table (prompts with more than 272K input tokens) | $50.00 per 1M units | 2026-10-25 | Pricing | OpenAI API |
| GPT-6 Luna | OpenAI | 12-hour cache writes 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) | $0.25 per 1M units | 2026-10-25 | Pricing | OpenAI API |
| GPT-6 Luna | OpenAI | 12-hour cache writes, > 272K input tokens 12-hour cache write, long-context column of the pricing page Standard table (prompts with more than 272K input tokens) | $0.50 per 1M units | 2026-10-25 | Pricing | OpenAI API |
| GPT-6 Sol | OpenAI | 12-hour cache writes 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) | $5.00 per 1M units | 2026-10-25 | Pricing | OpenAI API |
| GPT-6 Sol | OpenAI | 12-hour cache writes, > 272K input tokens 12-hour cache write, long-context column of the pricing page Standard table (prompts with more than 272K input tokens) | $10.00 per 1M units | 2026-10-25 | Pricing | OpenAI API |
| GPT-6.1 Sol | OpenAI | 12-hour cache writes 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) | $5.00 per 1M units | 2026-10-25 | Pricing | OpenAI API |
| GPT-6.1 Sol | OpenAI | 12-hour cache writes, > 272K input tokens 12-hour cache write, long-context column of the pricing page Standard table (prompts with more than 272K input tokens) | $10.00 per 1M units | 2026-10-25 | Pricing | OpenAI API |
| GPT-Image-2.5 Flare | OpenAI | Batch image input tokens Batch (pricing page, Batch table; 50% of Standard) | $4.00 per 1M units | 2026-10-25 | Pricing | OpenAI API |
| GPT-Image-2.5 Flare | OpenAI | Batch image output tokens Batch (pricing page, Batch table; 50% of Standard) | $15.00 per 1M units | 2026-10-25 | Pricing | OpenAI API |
| GPT-Image-2.5 Flare | OpenAI | Cached image input tokens Standard; the pricing page states cached input rates for GPT Image 2 and GPT Image 2.5 only apply to images generated with the Responses API | $2.00 per 1M units | 2026-10-25 | Pricing | OpenAI API |
| GPT-Image-2.5 Sunburst | OpenAI | Batch image input tokens Batch (pricing page, Batch table; 50% of Standard) | $4.00 per 1M units | 2026-10-25 | Pricing | OpenAI API |
| GPT-Image-2.5 Sunburst | OpenAI | Batch image output tokens Batch (pricing page, Batch table; 50% of Standard) | $15.00 per 1M units | 2026-10-25 | Pricing | OpenAI API |
| GPT-Image-2.5 Sunburst | OpenAI | Cached image input tokens Standard; the pricing page states cached input rates for GPT Image 2 and GPT Image 2.5 only apply to images generated with the Responses API | $2.00 per 1M units | 2026-10-25 | Pricing | OpenAI API |
| Grok Imagine Image | xAI | — Media input: per input image (e.g. image editing) | $0.002 per unit | 2026-10-25 | API Pricing |
| Grok Imagine Image 2.0 | xAI | — Media input: per input image (e.g. image editing) | $0.01 per unit | 2026-10-25 | API Pricing |
| Grok Imagine Image Quality | xAI | — Media input: per input image (e.g. image editing) | $0.01 per unit | 2026-10-25 | API Pricing |
| Nano Banana 2 Lite | — Batch image output tokens: Batch API price as listed on the pricing page (50% of Standard) | $15.00 per 1M units | 2026-10-25 | Gemini Developer API pricing | |
| Nano Banana 2.1 | — Batch image output tokens: Batch API price as listed on the pricing page (50% of Standard) | $15.00 per 1M units | 2026-10-25 | Gemini Developer API pricing | |
| Qwen-Image-3.0-Pro | Alibaba Cloud (Qwen) | — Per input image (image editing / reference input), same price for 1K and 2K tiers. | $0.003 per unit | 2026-10-25 | qwen-image-3.0-pro Model Info - Alibaba Cloud Model Studio |
| Qwen3.7-Plus | Alibaba Cloud (Qwen) | Explicit cache read Explicit cache read (explicit cache hit). | $0.04 per 1M units | 2026-10-25 | qwen3.7-plus Model Info - Alibaba Cloud Model Studio |
| Qwen3.7-Plus | Alibaba Cloud (Qwen) | Explicit cache read, > 256K input tokens Explicit cache read (explicit cache hit). | $0.12 per 1M units | 2026-10-25 | qwen3.7-plus Model Info - Alibaba Cloud Model Studio |
| Qwen3.8-2.4T-A95B | Alibaba Cloud (Qwen) | — Explicit cache hit. | $0.17 per 1M units | 2026-10-25 | qwen3.8-2.4t-a95b Model Info - Alibaba Cloud Model Studio |
| Qwen3.8-27B | Alibaba Cloud (Qwen) | — Explicit cache read (explicit cache hit). | $0.05 per 1M units | 2026-10-25 | qwen3.8-27b Model Info - Alibaba Cloud Model Studio |
| Qwen3.8-Flash | Alibaba Cloud (Qwen) | — Explicit cache read (explicit cache hit). | $0.016 per 1M units | 2026-10-25 | qwen3.8-flash Model Info - Alibaba Cloud Model Studio |
| Qwen3.8-Max | Alibaba Cloud (Qwen) | — Explicit cache read (explicit cache hit). | $0.17 per 1M units | 2026-10-25 | qwen3.8-max Model Info - Alibaba Cloud Model Studio |
Sorting happens within each currency; amounts are never converted between currencies. Rows show the base tier; "tiered" models charge different rates above an input size per request. Token tables for text include only models that produce text; image-only and speech-only models appear in their image and audio dimensions. Tool fees and other charges are listed with what they cover and never ranked against each other.