Models.Priced per token.

Anthropic claude-opus-5-5: the Opus tier of Claude, Anthropic's most capable line for complex reasoning, agents and software engineering. Text and image input.

Input

$4.00

Output

$20.00

Cached input

$0.200

Context1M

Anthropic claude-fable-5-1: the Fable tier above Opus in Anthropic's Claude 5 family, for the hardest reasoning, research and coding work. Text and image input.

Input

$10.00

Output

$50.00

Cached input

$0.250

Context1M

Z.ai GLM-5.3: GLM flagship chat and reasoning model with strong coding and agent performance.

Input

$1.18

Output

$4.13

Cached input

$0.295

Context1M

Anthropic claude-fable-5: the Fable tier above Opus in Anthropic's Claude 5 family, for the hardest reasoning, research and coding work. Text and image input.

Input

$10.00

Output

$50.00

Cached input

$1.00

Context1M

OpenAI gpt-5.5: a GPT-series chat and reasoning model available through the Chat Completions and Responses APIs. Text and image input.

Input

$5.00

Output

$30.00

Cached input

$0.500

Context922K

OpenAI gpt-5.4: a GPT-series chat and reasoning model available through the Chat Completions and Responses APIs. Text and image input.

Input

$2.50

Output

$15.00

Cached input

$0.250

Context—

Google gemini-3.1-flash-image-preview: Gemini image generation and editing. Priced per generated image.

Input

$0.500

Output

$60.00

Cached input

$0.050

Context128K

Z.ai GLM-5: GLM flagship chat and reasoning model with strong coding and agent performance.

Input

$0.580

Output

$2.61

Cached input

$0.145

Context200K

Anthropic claude-opus-4.6: the Opus tier of Claude, Anthropic's most capable line for complex reasoning, agents and software engineering. Text and image input.

Input

$5.00

Output

$25.00

Cached input

$0.500

Context1M

ByteDance doubao-seedream-4-5: Seedream image generation. Priced per image.

Input

$32.00

Output

$32.00

Cached input

$3.20

Context—

Anthropic's most capable Claude 4.5 model for complex reasoning, software engineering and agentic workflows. Vision input, 200K context window, up to 64K output tokens. Released 2025-11.

Input

$5.00

Output

$25.00

Cached input

$0.500

Context200K

Google gemini-3-pro-image-preview: Gemini image generation and editing. Priced per generated image.

Input

$2.00

Output

$120.00

Cached input

$0.200

Context64K

Anthropic's fastest and most affordable Claude 4.5 model. Near-frontier coding and agentic performance at a fraction of the cost, with vision input, a 200K context window and up to 64K output tokens. Released 2025-10.

Input

$1.00

Output

$5.00

Cached input

$0.100

Context200K

Anthropic's balanced Claude 4.5 model: strong at coding, long-running agents and computer use, with vision input, a 200K context window and up to 64K output tokens. Released 2025-09-29.

Input

$3.00

Output

$15.00

Cached input

$0.300

Context200K

ByteDance doubao-seedance-2-0-260128: Seedance video generation. Priced per generation.

Input

$6.78

Output

$6.78

Cached input

$6.78

Context—

ByteDance doubao-seedance-2-5-260628: Seedance video generation. Priced per generation.

Input

$10.32

Output

$10.32

Cached input

$10.32

Context—

ByteDance doubao-seedream-5-0-260128: Seedream image generation. Priced per image.

per request

$0.044

Context—

ByteDance doubao-seedream-5-0-pro-260628: Seedream image generation. Priced per image.

per request

$0.044

Context—

Alibaba qwen3-max: Qwen general-purpose chat and coding model.

Input

$0.362

Output

$1.45

Cached input

$0.036

Context262K

Z.ai GLM-5.3-Flash: fast, low-cost GLM tier for high-throughput chat and coding.

Input

$0.118

Output

$0.413

Cached input

$0.034

Context—

OpenAI gpt-6.1-sol: a GPT-series chat and reasoning model available through the Chat Completions and Responses APIs. Text and image input.

Input

$2.00

Output

$10.00

Cached input

$0.100

Context1M

Anthropic claude-sonnet-5-5: the Sonnet tier of Claude, balancing capability, speed and cost for coding and everyday agent workloads. Text and image input.

Input

$2.00

Output

$10.00

Cached input

$0.200

Context1M

OpenAI gpt-6-luna: a GPT-series chat and reasoning model available through the Chat Completions and Responses APIs. Text and image input.

Input

$0.100

Output

$0.500

Cached input

$0.010

Context1M

OpenAI gpt-6-sol: a GPT-series chat and reasoning model available through the Chat Completions and Responses APIs. Text and image input.

Input

$2.00

Output

$10.00

Cached input

$0.200

Context1M

DeepSeek deepseek-flash: the fast, low-cost DeepSeek tier with reasoning tokens reported separately.

Input

$0.147

Output

$0.590

Cached input

$0.0029

Context1M

OpenAI gpt-image-2.5-flare: image generation and editing model. Priced per generated image.

Input

$5.00

Output

$30.00

Cached input

$1.25

Context—

OpenAI gpt-image-2.5-sunburst: image generation and editing model. Priced per generated image.

Input

$5.00

Output

$30.00

Cached input

$1.25

Context—

OpenAI gpt-6-astra: a GPT-series chat and reasoning model available through the Chat Completions and Responses APIs. Text and image input.

Input

$10.00

Output

$50.00

Cached input

$1.00

Context1M

Google gemini-3.8-flash: the fast, price-efficient Gemini Flash tier with long context and optional thinking. Text and image input.

Input

$0.750

Output

$3.75

Cached input

$0.075

Context1M

Google gemini-3.7-flash: the fast, price-efficient Gemini Flash tier with long context and optional thinking. Text and image input.

Input

$0.750

Output

$3.75

Cached input

$0.075

Context1M

Alibaba qwen3.8-max: Qwen general-purpose chat and coding model.

Input

$1.77

Output

$5.31

Cached input

$0.221

Context1M

Anthropic claude-opus-5: the Opus tier of Claude, Anthropic's most capable line for complex reasoning, agents and software engineering. Text and image input.

Input

$5.00

Output

$25.00

Cached input

$0.500

Context—

Moonshot AI Kimi-K3: Kimi frontier chat model with long context and strong agentic tool use. Text and image input.

Input

$2.95

Output

$14.75

Cached input

$0.295

Context1M

OpenAI gpt-5.6-sol: a GPT-series chat and reasoning model available through the Chat Completions and Responses APIs. Text and image input.

Input

$4.00

Output

$20.00

Cached input

$0.400

Context—

Google gemini-3.1-flash-lite-image: Gemini image generation and editing. Priced per generated image.

Input

$0.250

Output

$30.00

Cached input

$0.025

Context—

Z.ai GLM-5.2: GLM flagship chat and reasoning model with strong coding and agent performance.

Input

$1.18

Output

$4.13

Cached input

$0.295

Context1M

Moonshot AI Kimi-K2.7-Code: Kimi variant tuned for software engineering and agentic coding.

Input

$0.942

Output

$3.91

Cached input

$0.188

Context262K

MiniMax MiniMax-M3: MiniMax M-series model for agentic coding and long-horizon tasks.

Input

$0.304

Output

$1.22

Cached input

$0.061

Context1M

Anthropic claude-opus-4-8: the Opus tier of Claude, Anthropic's most capable line for complex reasoning, agents and software engineering. Text and image input.

Input

$5.00

Output

$25.00

Cached input

$0.500

Context1M

Google gemini-3.5-flash: the fast, price-efficient Gemini Flash tier with long context and optional thinking. Text and image input.

Input

$1.50

Output

$9.00

Cached input

$0.150

Context1M

Alibaba qwen3.7-max: Qwen general-purpose chat and coding model.

Input

$1.74

Output

$5.22

Cached input

$0.348

Context1M

Qwen (Alibaba) happyhorse-1.0-i2v: video generation / editing model. Priced per generation.

per request

$0.13

Context—

Qwen (Alibaba) happyhorse-1.0-r2v: video generation / editing model. Priced per generation.

per request

$0.13

Context—

Qwen (Alibaba) happyhorse-1.0-t2v: video generation / editing model. Priced per generation.

per request

$0.13

Context—

DeepSeek deepseek-v4-flash: the fast, low-cost DeepSeek tier with reasoning tokens reported separately.

Input

$0.147

Output

$0.590

Cached input

$0.0029

Context1M

DeepSeek deepseek-v4-pro: general-purpose DeepSeek chat model with strong coding and reasoning at low cost.

Input

$0.664

Output

$1.99

Cached input

$0.022

Context1M

OpenAI gpt-image-2: image generation and editing model. Priced per generated image.

Input

$5.00

Output

$30.00

Cached input

$1.25

Context—

Moonshot AI Kimi-K2.6: Kimi frontier chat model with long context and strong agentic tool use. Text and image input.

Input

$0.942

Output

$3.91

Cached input

$0.159

Context256K

Anthropic claude-opus-4-7: the Opus tier of Claude, Anthropic's most capable line for complex reasoning, agents and software engineering. Text and image input.

Input

$5.00

Output

$25.00

Cached input

$0.500

Context1M

Z.ai GLM-5-Turbo: fast, low-cost GLM tier for high-throughput chat and coding.

Input

$0.725

Output

$3.19

Cached input

$0.181

Context200K

Z.ai GLM-5V-Turbo: GLM vision-language model for image understanding and multimodal chat.

Input

$0.725

Output

$3.19

Cached input

$0.181

Context200K

Z.ai GLM-5.1: GLM flagship chat and reasoning model with strong coding and agent performance.

Input

$0.885

Output

$3.54

Cached input

$0.192

Context200K

Alibaba qwen3.6-plus: Qwen general-purpose chat and coding model.

Input

$0.290

Output

$1.74

Cached input

$0.058

Context1M

Alibaba qwen3.5-plus: Qwen general-purpose chat and coding model.

Input

$0.116

Output

$0.696

Cached input

$0.023

Context1M

MiniMax MiniMax-M2.7: MiniMax M-series model for agentic coding and long-horizon tasks.

Input

$0.304

Output

$1.22

Cached input

$0.061

Context200K

Google gemini-3.1-flash-lite-preview: the lightest, lowest-latency Gemini tier for high-volume, cost-sensitive workloads. Text and image input.

Input

$0.250

Output

$1.50

Cached input

$0.025

Context1M

Alibaba qwen3.5-flash: fast, low-cost Qwen tier for high-volume chat.

Input

$0.029

Output

$0.290

Cached input

$0.0058

Context1M

Anthropic claude-sonnet-4.6: the Sonnet tier of Claude, balancing capability, speed and cost for coding and everyday agent workloads. Text and image input.

Input

$3.00

Output

$15.00

Cached input

$0.300

Context1M

Alibaba Qwen3.5-397B-A17B: Qwen general-purpose chat and coding model.

Input

$0.174

Output

$1.04

Cached input

$0.017

Context262K

MiniMax MiniMax-M2.5: MiniMax M-series model for agentic coding and long-horizon tasks.

Input

$0.304

Output

$1.22

Cached input

$0.030

Context200K

Moonshot AI Kimi-K2.5: Kimi frontier chat model with long context and strong agentic tool use. Text and image input.

Input

$0.580

Output

$3.04

Cached input

$0.101

Context256K

Alibaba Qwen3-Max-Thinking: Qwen reasoning variant with extended thinking (reasoning tokens billed as output).

Input

$0.362

Output

$1.45

Cached input

$0.362

Context252K

Google gemini-3-flash-preview: the fast, price-efficient Gemini Flash tier with long context and optional thinking. Text and image input.

Input

$0.500

Output

$3.00

Cached input

$0.050

Context1M

DeepSeek DeepSeek-V3.2: general-purpose DeepSeek chat model with strong coding and reasoning at low cost.

Input

$0.288

Output

$0.432

Cached input

$0.029

Context128K

Google gemini-3-pro-preview: the Gemini Pro tier for complex reasoning, coding and long-context analysis. Text and image input.

Input

$2.00

Output

$12.00

Cached input

$0.200

Context1M

Google gemini-2.5-flash-image: Gemini image generation and editing. Priced per generated image.

Input

$0.300

Output

$30.00

Cached input

$0.030

Context32K

DeepSeek R1: open-weight reasoning model that emits chain-of-thought (billed as reasoning tokens) before answering. Strong on math and code.

Input

$0.578

Output

$2.31

Cached input

$0.058

Context128K

Anthropic claude-sonnet-5: the Sonnet tier of Claude, balancing capability, speed and cost for coding and everyday agent workloads. Text and image input.

Input

$2.00

Output

$10.00

Cached input

$0.200

Context—

ByteDance doubao-seed-1-8-251228: Doubao Seed chat model with text and image input.

Input

$0.116

Output

$0.290

Cached input

$0.012

Context—

ByteDance doubao-seed-2-1-pro-260628: Doubao Seed chat model with text and image input.

Input

$0.885

Output

$4.42

Cached input

$0.177

Context—

ByteDance doubao-seed-2-1-turbo-260628: Doubao Seed chat model with text and image input.

Input

$0.442

Output

$2.21

Cached input

$0.088

Context—

Google's fast, price-efficient Gemini 2.5 workhorse with a 1M-token context window, up to 65K output tokens and optional thinking. Accepts text and images. Released 2025-06.

Input

$0.300

Output

$2.50

Cached input

$0.030

Context1M

Google gemini-3.1-flash-lite: the lightest, lowest-latency Gemini tier for high-volume, cost-sensitive workloads. Text and image input.

Input

$0.250

Output

$1.50

Cached input

$0.025

Context—

Google gemini-3.1-pro-preview: the Gemini Pro tier for complex reasoning, coding and long-context analysis. Text and image input.

Input

$2.00

Output

$12.00

Cached input

$0.200

Context—

Google omni-fast: low-latency multimodal model for audio and voice-style interactions.

per request

$0.080

Context—

Google omni-fast-v2v: low-latency multimodal model for audio and voice-style interactions.

per request

$0.12

Context—

OpenAI gpt-5.6-luna: a GPT-series chat and reasoning model available through the Chat Completions and Responses APIs. Text and image input.

Input

$0.200

Output

$1.20

Cached input

$0.020

Context—

OpenAI gpt-5.6-terra: a GPT-series chat and reasoning model available through the Chat Completions and Responses APIs. Text and image input.

Input

$2.00

Output

$12.00

Cached input

$0.200

Context—

Alibaba qwen-deep-research: agentic deep-research model that plans, browses and synthesises long reports.

Input

$7.83

Output

$23.62

Cached input

$7.83

Context1M

Alibaba qwen-flash: fast, low-cost Qwen tier for high-volume chat.

Input

$0.021

Output

$0.214

Cached input

$0.0021

Context1M

Alibaba qwen3-vl-flash: Qwen vision-language model for image understanding and multimodal chat.

Input

$0.022

Output

$0.216

Cached input

$0.0022

Context262K

Alibaba qwen3-vl-plus: Qwen vision-language model for image understanding and multimodal chat.

Input

$0.145

Output

$1.45

Cached input

$0.015

Context262K

Alibaba's Qwen text-embedding-v4: multilingual embeddings returning 1024-dimensional vectors (verified through NXIO on 2026-09-28). Billed on input tokens only.

Input

$0.071

Output

$0

Cached input

—

Context8K

Qwen (Alibaba) wan2.6-i2v: video generation / editing model. Priced per generation.

per request

$0.086

Context—

Qwen (Alibaba) wan2.6-t2v: video generation / editing model. Priced per generation.

per request

$0.086

Context—

Qwen (Alibaba) wan2.7-i2v: video generation / editing model. Priced per generation.

per request

$0.086

Context—

Alibaba wan2.7-image: Wan image generation. Priced per image.

per request

$0.029

Context—

Alibaba wan2.7-image-pro: Wan image generation. Priced per image.

per request

$0.072

Context—

Qwen (Alibaba) wan2.7-r2v: video generation / editing model. Priced per generation.

per request

$0.086

Context—

Qwen (Alibaba) wan2.7-t2v: video generation / editing model. Priced per generation.

per request

$0.086

Context—

Qwen (Alibaba) wan2.7-videoedit: video generation / editing model. Priced per generation.

per request

$0.086

Context—

Qwen (Alibaba) wan3.0-video: video generation / editing model. Priced per generation.

per request

$0.044

Context—

Qwen (Alibaba) wan3.0-video-prime: video generation / editing model. Priced per generation.

per request

$0.066

Context—

Skywork (Kunlun) skyreels: text in, video out, priced per request, available through the NXIO gateway.

per request

$0.080

Context—

xAI grok-imagine-video-1.5-preview: video generation / editing model. Priced per generation.

per request

$0.080

Context—