跳到正文
This page is also in English.
Switch to English
API Pricing

5,032 个接口 · 11,852 个价格 · 最近采集 9 分钟前 · 30 天内 442 条需关注的调价

每个平台公布的每一个价格

采集器定时抓取 11 个平台公开的全部价格(默认每日一次),并完整保存原始页面快照,绝无任何手工篡改。搜索具体模型,即可查看当前各单位下的最低价格。

第 1201–1300 条,共 1,867 条

“最低”是本次搜索在该单位下的最低精确价。不同行的分辨率、时长或质量可能不同,请看规格说明,或把规格加进搜索(例如“1080”)。

每百万输出 token70

平台模型 · 计价项美元平台原文生效于
OpenRouter未披露NVIDIA: Nemotron 3 Ultravia KT$2.2/百万输出completion: 0.0000022 / token (endpoint kt)
OpenRouter官方OpenAI: o3 Mini (batch)$2.2/百万输出completion: 0.0000022 / token (endpoint openai)
APIMart未披露gemini-3-flash-preview$2.4/百万输出Input $0.4 / Output $2.4 · Per 1M tokens
APIMart未披露gemini-3-flash-preview-nothinking$2.4/百万输出Input $0.4 / Output $2.4 · Per 1M tokens
APIMart未披露gemini-3-flash-preview-thinking$2.4/百万输出Input $0.4 / Output $2.4 · Per 1M tokens
Runware未披露MiniMax M3> 512K context$2.4/百万输出$0.0000024 / outputToken · > 512K context
OpenRouter第三方Moondream: Moondream 3 Preview$2.5/百万输出completion: 0.0000025 / token (endpoint primeintellect/fp8)
OpenRouter官方Qwen: Qwen3 Coder Flash提示词超过 128K$2.6/百万输出completion: 0.0000026 / token (>128000; endpoint alibaba)
APIMart未披露MiniMax-H3-Context-IR$2.629/百万输出26.2865 ~$2.6287平台声称官方价 $3.286/百万输出会员价: platinum $2.497 · diamond $2.366
Replicate平台自部署meta/meta-llama-3-70b$2.75/百万输出$2.75 per million output tokens
Replicate平台自部署meta/meta-llama-3-70b-instruct$2.75/百万输出$2.75 per million output tokens
Aiberm未披露Gemini 3 Pro Preview$2.76/百万输出model_ratio 1 × $2/M · group default ×0.23 · completion_ratio 6平台声称官方价 $12/百万输出
Aiberm未披露Gemini 3 Pro Preview Thinking$2.76/百万输出model_ratio 1 × $2/M · group default ×0.23 · completion_ratio 6平台声称官方价 $12/百万输出
Aiberm未披露Gemini 3 Pro$2.76/百万输出model_ratio 1 × $2/M · group default ×0.23 · completion_ratio 6平台声称官方价 $12/百万输出
Aiberm未披露Gemini 3 Pro Mcpmark$2.76/百万输出model_ratio 1 × $2/M · group default ×0.23 · completion_ratio 6平台声称官方价 $12/百万输出
OpenRouter官方Google: Gemini 3 Flash Previewvia Google AI Studio$3/百万输出completion: 0.000003 / token (endpoint google-ai-studio)
OpenRouter官方Google: Gemini 3 Flash Preview推理via Google AI Studio$3/百万输出internal_reasoning: 0.000003 / token (endpoint google-ai-studio)
Replicate官方合作google/gemini-3-flash$3/百万输出$3 per million output tokens
Runware未披露Gemini 3 Flash$3/百万输出$0.000003 / outputToken
Fal平台自部署Moondream3 Preview [Caption]$3.5/百万输出Your request will cost $0.40 per million input tokens, and $3.50 per million output tokens.
Fal平台自部署Moondream3 Preview [Detect]$3.5/百万输出Your request will cost $0.40 per million input tokens, and $3.50 per million output tokens.
Fal平台自部署Moondream3 Preview [Point]$3.5/百万输出Your request will cost $0.40 per million input tokens, and $3.50 per million output tokens.
Fal平台自部署Moondream 3 Preview [Query]$3.5/百万输出Your request will cost $0.40 per million input tokens, and $3.50 per million output tokens.
Fal平台自部署Moondream3 Preview [Segment]$3.5/百万输出Your request will cost $0.40 per million input tokens, and $3.50 per million output tokens.
KIE未披露Gemini 3 Pro$3.5/百万输出700 credits per million tokens ($3.5)平台声称官方或 Fal 价 $12/百万输出
OpenRouter第三方Qwen: Qwen3 235B A22B Thinking 2507$3.5/百万输出completion: 0.0000035 / token (endpoint venice/fp8)
APIMart未披露o3-mini$3.52/百万输出Input $0.88 / Output $3.52 · Per 1M tokens
APIMart未披露o3-mini-2025-01-31$3.52/百万输出Input $0.88 / Output $3.52 · Per 1M tokens
APIMart未披露o3-mini-2025-01-31-high$3.52/百万输出Input $0.88 / Output $3.52 · Per 1M tokens
APIMart未披露o3-mini-2025-01-31-low$3.52/百万输出Input $0.88 / Output $3.52 · Per 1M tokens
APIMart未披露o3-mini-2025-01-31-medium$3.52/百万输出Input $0.88 / Output $3.52 · Per 1M tokens
APIMart未披露o3-mini-high$3.52/百万输出Input $0.88 / Output $3.52 · Per 1M tokens
APIMart未披露o3-mini-low$3.52/百万输出Input $0.88 / Output $3.52 · Per 1M tokens
APIMart未披露o3-mini-medium$3.52/百万输出Input $0.88 / Output $3.52 · Per 1M tokens
OpenRouter官方MiniMax: MiniMax H3 Image$3.8/百万输出output_image: 0.0000038 USD / token
Runware未披露MiniMax H3 ImageImage output$3.8/百万输出$0.0000038 / outputToken · Image output
OpenRouter官方OpenAI: o3 (batch)$4/百万输出completion: 0.000004 / token (endpoint openai)
Aiberm未披露Kimi K3$5.29/百万输出model_ratio 2.3 × $2/M · group default ×0.23 · completion_ratio 5平台声称官方价 $15/百万输出
APIMart未披露o3$6.4/百万输出Input $1.6 / Output $6.4 · Per 1M tokens
APIMart未披露o3-2025-04-16$6.4/百万输出Input $1.6 / Output $6.4 · Per 1M tokens
OpenRouter官方MiniMax: MiniMax H3 Image Highspeed$7.6/百万输出output_image: 0.0000076 USD / token
Runware未披露MiniMax H3 Image HighspeedImage output$7.6/百万输出$0.0000076 / outputToken · Image output
APIMart未披露gpt-4-32k$8/百万输出Input $2 / Output $8 · Per 1M tokens
APIMart未披露gpt-4-32k-0613$8/百万输出Input $2 / Output $8 · Per 1M tokens
OpenRouter官方OpenAI: o3$8/百万输出completion: 0.000008 / token (endpoint openai)
APIMart未披露gemini-3-pro-preview$9.6/百万输出Input $1.6 / Output $9.6 · Per 1M tokens
APIMart未披露gemini-3-pro-preview-thinking$9.6/百万输出Input $1.6 / Output $9.6 · Per 1M tokens
Runware未披露Gemini 4 ArgonThrough Nov 30 · 2026$10/百万输出$0.00001 / outputToken · Through Nov 30, 2026
OpenRouter未披露MoonshotAI: Kimi K3via Relace · fp4$11.006/百万输出completion: 0.0000110055 / token (endpoint relace/fp4)
OpenRouter第三方MoonshotAI: Kimi K3 (batch)$11.4/百万输出completion: 0.0000114 / token (endpoint deepinfra/bf16)
APIMart未披露kimi-k3$12/百万输出Input $2.4 / Output $12 · Per 1M tokens
KIE未披露Kimi K3$12/百万输出2400 credits per million tokens ($12.00)平台声称官方或 Fal 价 $15/百万输出
OpenRouter官方Google: Nano Banana Pro (Gemini 3 Pro Image)Text$12/百万输出completion: 0.000012 / token
OpenRouter官方Google: Nano Banana Pro (Gemini 3 Pro Image)推理$12/百万输出internal_reasoning: 0.000012 / token
OpenRouter官方Google: Nano Banana Pro (Gemini 3 Pro Image Preview)Text$12/百万输出completion: 0.000012 / token
OpenRouter官方Google: Nano Banana Pro (Gemini 3 Pro Image Preview)推理$12/百万输出internal_reasoning: 0.000012 / token
Replicate官方合作google/gemini-3-pro≤ 200 · 000$12/百万输出$0.012 per thousand output tokens (input token ≤ 200,000)
Runware未披露Kimi K3$15/百万输出$0.000015 / outputToken
Replicate官方合作google/gemini-3-pro> 200 · 000$18/百万输出$0.018 per thousand output tokens (input token > 200,000)
APIMart未披露o3-deep-research$32/百万输出Input $8 / Output $32 · Per 1M tokens
APIMart未披露o3-deep-research-2025-06-26$32/百万输出Input $8 / Output $32 · Per 1M tokens
OpenRouter官方Google: Nano Banana Pro (Gemini 3 Pro Image Preview)via Google AI Studio · global/flex$60/百万输出output_image: 0.00006 USD / token
APIMart未披露o3-pro$64/百万输出Input $16 / Output $64 · Per 1M tokens
APIMart未披露o3-pro-2025-06-10$64/百万输出Input $16 / Output $64 · Per 1M tokens
OpenRouter官方OpenAI: o3 Pro$80/百万输出completion: 0.00008 / token (endpoint openai)
OpenRouter官方Google: Nano Banana Pro (Gemini 3 Pro Image)via Google AI Studio$120/百万输出output_image: 0.00012 USD / token
OpenRouter官方Google: Nano Banana Pro (Gemini 3 Pro Image)via Google Vertex$120/百万输出output_image: 0.00012 USD / token
OpenRouter官方Google: Nano Banana Pro (Gemini 3 Pro Image)$120/百万输出image_output: 0.00012 / token
OpenRouter官方Google: Nano Banana Pro (Gemini 3 Pro Image Preview)via Google AI Studio · global$120/百万输出output_image: 0.00012 USD / token
OpenRouter官方Google: Nano Banana Pro (Gemini 3 Pro Image Preview)$120/百万输出image_output: 0.00012 / token

每百万 token4

平台模型 · 计价项美元平台原文生效于
Fal平台自部署Nemotron 3 Nano Omni$6/百万 token最低$0.006 / 1000 tokens
Fal平台自部署Nemotron 3 Nano Omni$6/百万 token最低$0.006 / 1000 tokens
Fal平台自部署Nemotron 3 Nano Omni$10/百万 token$0.01 / 1000 tokens
Fal平台自部署Nemotron 3 Nano Omni$10/百万 token$0.01 / 1000 tokens

每百万缓存读取 token26

平台模型 · 计价项美元平台原文生效于
OpenRouter官方Mistral: Ministral 3 3B 2512via Mistral · zdr$0.01/百万缓存读最低input_cache_read: 0.00000001 / token (endpoint mistral/zdr)
OpenRouter第三方NVIDIA: Nemotron 3 Nano 30B A3B$0.012/百万缓存读input_cache_read: 0.000000012 / token (endpoint nebius/fp8)
Runware未披露Gemma 4 31BPer 1M$0.012/百万缓存读$1.2e-8 / cachedInputToken
OpenRouter官方Mistral: Ministral 3 8B 2512via Mistral · zdr$0.015/百万缓存读input_cache_read: 0.000000015 / token (endpoint mistral/zdr)
OpenRouter官方Mistral: Ministral 3 14B 2512via Mistral · zdr$0.02/百万缓存读input_cache_read: 0.00000002 / token (endpoint mistral/zdr)
OpenRouter第三方Tencent: Hy3off-peak 1600–0000 UTC · via Tencent · fp8$0.021/百万缓存读input_cache_read: 0.000000020625 / token (utc_start 1600, utc_end 0; endpoint tencent/fp8)
OpenRouter第三方Meta: Muse Glimmer 30B (batch)$0.032/百万缓存读input_cache_read: 0.000000032 / token (endpoint deepinfra/bf16)
OpenRouter第三方Tencent: Hy3peak 0000–1600 UTC · via Tencent · fp8$0.033/百万缓存读input_cache_read: 0.000000033 / token (utc_start 0, utc_end 1600; endpoint tencent/fp8)
OpenRouter官方Qwen: Qwen3 Coder Flash$0.039/百万缓存读input_cache_read: 0.000000039 / token (endpoint alibaba)
OpenRouter第三方Meta: Muse Glimmer 30B$0.04/百万缓存读input_cache_read: 0.00000004 / token (endpoint deepinfra/bf16)
OpenRouter官方Mistral: Mistral Medium 3via Mistral · zdr$0.04/百万缓存读input_cache_read: 0.00000004 / token (endpoint mistral/zdr)
OpenRouter第三方MiniMax: MiniMax M3 (batch)$0.045/百万缓存读input_cache_read: 0.0000000448 / token (endpoint deepinfra/fp8)
Runware未披露Shieldstral 1.0 3BPer 1M$0.045/百万缓存读$4.5e-8 / cachedInputToken
OpenRouter官方Google: Gemini 3 Flash Previewvia Google AI Studio$0.05/百万缓存读input_cache_read: 0.00000005 / token (endpoint google-ai-studio)
OpenRouter第三方Google: Gemma 3 12Bvia NextBit · int4$0.05/百万缓存读input_cache_read: 0.00000005 / token (endpoint nextbit/int4)
OpenRouter第三方MiniMax: MiniMax M3via CoreWeave · fp4$0.05/百万缓存读input_cache_read: 0.00000005 / token (endpoint coreweave/fp4)
OpenRouter官方Mistral: Mistral Large 3 2512via Mistral · zdr$0.05/百万缓存读input_cache_read: 0.00000005 / token (endpoint mistral/zdr)
OpenRouter未披露Motif Technologies: Motif 3$0.06/百万缓存读input_cache_read: 0.00000006 / token (endpoint kt)
OpenRouter第三方Tencent: Hy3 preview$0.06/百万缓存读input_cache_read: 0.00000006 / token (endpoint gmicloud/bf16)
Runware未披露MiniMax M3≤ 512K context$0.06/百万缓存读$6e-8 / cachedInputToken · ≤ 512K context
OpenRouter官方Qwen: Qwen3 Coder Flash提示词超过 32K$0.065/百万缓存读input_cache_read: 0.000000065 / token (>32000; endpoint alibaba)
OpenRouter第三方Qwen: Qwen3 Coder Next$0.07/百万缓存读input_cache_read: 0.00000007 / token (endpoint parasail/bf16)
Aiberm未披露o3 Mini$0.076/百万缓存读model_ratio 0.55 × $2/M · group default ×0.23 · cache_ratio 0.3
OpenRouter官方Google: Gemini 3 Flash Previewvia Google AI Studio$0.1/百万缓存读input_audio_cache: 0.0000001 / token (endpoint google-ai-studio)
OpenRouter未披露Google: Gemma 4 31Bvia KT$0.1/百万缓存读input_cache_read: 0.0000001 / token (endpoint kt)
OpenRouter未披露NVIDIA: Nemotron 3 Ultravia KT$0.1/百万缓存读input_cache_read: 0.0000001 / token (endpoint kt)