AI 正在成为一个全球市场。
HotON.ai 帮你读懂它。
模型在变化,价格在变化,算力成本在变化。HotON.ai 把分散的 AI 信号整理成结构化的市场情报。
模型价格追踪
追踪主流 AI 模型的 API 价格、Token 成本、上下文长度与价格变化。
AI 市场热度
实时发现哪些模型、厂商、地区与类别正在升温。
推理成本情报
比较不同模型、地区与基础设施条件下运行 AI 任务的真实成本。
模型效率指数
不只看价格,还看速度、稳定性、输出质量与每次成功任务的成本。
全球 AI 市场的实时视图
从模型价格、使用趋势到可用性、延迟与市场热度——整个 AI 经济一屏尽览。
热门模型
实时价格异动
24小时延迟监测
tok/s可用性信号
实时最新模型发布
最近类别热度
各类别动能追踪智能的价格
AI 价格已不再简单。HotON.ai 帮助开发者、企业与投资者看清模型成本在不同厂商和类别间的变化。
模型价格榜
美元 / 百万 Token| 模型 | 地区 | 输入 | 输出 | 上下文 | Δ |
|---|---|---|---|---|---|
| DEEPSEEK-V4-FL DeepSeek | CN | $0.10 | $0.20 | 1049K | −26.3% |
| DEEPSEEK-V4-PR DeepSeek | CN | $0.44 | $0.87 | 1049K | +4.6% |
| QWEN3.6-FLASH Alibaba | CN | $0.19 | $1.13 | 1000K | −24.9% |
| NEMOTRON-3-SUP NVIDIA | US | $0.09 | $0.45 | 1000K | −23.9% |
| LLAMA-4-MAVERI Meta | US | $0.15 | $0.60 | 1049K | −11.9% |
| LLAMA-4-SCOUT Meta | US | $0.08 | $0.30 | 10000K | +5.8% |
| GEMINI-3.1-FLA Google | US | $0.25 | $1.50 | 1049K | −10.0% |
| GEMINI-2.5-FLA Google | US | $0.10 | $0.40 | 1049K | −0.8% |
| MINIMAX-01 MiniMax | CN | $0.20 | $1.10 | 1000K | +8.0% |
| MIMO-V2.5 Xiaomi | CN | $0.14 | $0.28 | 1049K | −5.8% |
| MIMO-V2.5-PRO Xiaomi | CN | $0.44 | $0.87 | 1049K | +2.6% |
| MINIMAX-M3 MiniMax | CN | $0.30 | $1.20 | 1049K | −6.1% |
| QWEN3.5-PLUS-2 Alibaba | CN | $0.30 | $1.80 | 1000K | −6.0% |
| NOVA-2-LITE-V1 Amazon | US | $0.30 | $2.50 | 1000K | −31.4% |
| GEMINI-2.5-FLA Google | US | $0.30 | $2.50 | 1049K | −9.5% |
| GROK-4.3 xAI | US | $1.25 | $2.50 | 1000K | −8.7% |
| QWEN3.6-PLUS Alibaba | CN | $0.33 | $1.95 | 1000K | −31.6% |
| NEMOTRON-3-ULT NVIDIA | US | $0.50 | $2.50 | 1000K | −25.9% |
| QWEN3.7-PLUS Alibaba | CN | $0.40 | $1.60 | 1000K | −21.4% |
| MINIMAX-M1 MiniMax | CN | $0.40 | $2.20 | 1000K | +11.9% |
| PALMYRA-X5 Writer | US | $0.60 | $6.00 | 1040K | −26.9% |
| QWEN3.7-MAX Alibaba | CN | $1.25 | $3.75 | 1000K | −3.1% |
| GEMINI-3.5-FLA Google | US | $1.50 | $9.00 | 1049K | −8.3% |
| GEMINI-2.5-PRO Google | US | $1.25 | $10.00 | 1049K | +7.2% |
| GPT-5.4-NANO OpenAI | US | $0.20 | $1.25 | 400K | +10.0% |
| NOVA-LITE-V1 Amazon | US | $0.06 | $0.24 | 300K | −28.9% |
| KIMI-K2.5 Moonshot | CN | $0.40 | $1.90 | 262K | +9.8% |
| MINISTRAL-14B- Mistral | EU | $0.20 | $0.20 | 262K | −17.7% |
| MINIMAX-M2 MiniMax | CN | $0.26 | $1.00 | 205K | +2.9% |
| RING-2.6-1T InclusionAI | CN | $0.08 | $0.63 | 262K | −28.2% |
| MISTRAL-SMALL- Mistral | EU | $0.15 | $0.60 | 262K | −30.3% |
| MIMO-V2-FLASH Xiaomi | CN | $0.10 | $0.30 | 262K | −18.3% |
| MINIMAX-M2.1 MiniMax | CN | $0.29 | $0.95 | 205K | +2.6% |
| SEED-2.0-LITE ByteDance | CN | $0.25 | $2.00 | 262K | −3.3% |
| QWEN3.5-9B Alibaba | CN | $0.04 | $0.15 | 262K | +6.0% |
| LING-2.6-FLASH InclusionAI | CN | $0.01 | $0.03 | 262K | −15.6% |
| SEED-2.0-MINI ByteDance | CN | $0.10 | $0.40 | 262K | +5.2% |
| CODESTRAL-2508 Mistral | EU | $0.30 | $0.90 | 256K | −24.2% |
| SEED-1.6 ByteDance | CN | $0.25 | $2.00 | 262K | −27.8% |
| QWEN3.6-35B-A3 Alibaba | CN | $0.14 | $1.00 | 262K | −11.6% |
| STEP-3.5-FLASH StepFun | CN | $0.09 | $0.30 | 262K | −21.8% |
| NEMOTRON-3-NAN NVIDIA | US | $0.05 | $0.20 | 262K | +0.6% |
| SEED-1.6-FLASH ByteDance | CN | $0.08 | $0.30 | 262K | −2.0% |
| GLM-4.7-FLASH Zhipu | CN | $0.06 | $0.40 | 203K | −20.7% |
| NOVA-PREMIER-V Amazon | US | $2.50 | $12.50 | 1000K | −28.8% |
| LING-2.6-1T InclusionAI | CN | $0.08 | $0.63 | 262K | −16.0% |
| STEP-3.7-FLASH StepFun | CN | $0.20 | $1.15 | 256K | −7.4% |
| MISTRAL-LARGE- Mistral | EU | $0.50 | $1.50 | 262K | −30.5% |
| GEMMA-4-26B-A4 Google | US | $0.06 | $0.33 | 262K | +8.8% |
| GEMMA-4-31B-IT Google | US | $0.12 | $0.36 | 262K | −7.9% |
| LLAMA-3.3-70B- Meta | US | $0.10 | $0.32 | 131K | −8.9% |
| GRANITE-4.1-8B IBM | US | $0.05 | $0.10 | 131K | +7.4% |
| QWEN3.6-27B Alibaba | CN | $0.29 | $3.20 | 262K | −10.4% |
| KIMI-K2.6 Moonshot | CN | $0.68 | $3.42 | 262K | −7.3% |
| MINIMAX-M2.7 MiniMax | CN | $0.28 | $1.20 | 205K | −7.2% |
| GPT-5.4-MINI OpenAI | US | $0.75 | $4.50 | 400K | +9.6% |
| LFM-2-24B-A2B Liquid | US | $0.03 | $0.12 | 128K | +6.3% |
| CLAUDE-SONNET- Anthropic | US | $3.00 | $15.00 | 1000K | −12.9% |
| MINIMAX-M2.5 MiniMax | CN | $0.15 | $1.15 | 205K | +10.7% |
| SOLAR-PRO-3 Upstage | US | $0.15 | $0.60 | 128K | −9.4% |
| GLM-4.6V Zhipu | CN | $0.30 | $0.90 | 131K | −4.2% |
| TRINITY-MINI Arcee | US | $0.05 | $0.15 | 131K | −8.2% |
| DEEPSEEK-V3.2 DeepSeek | CN | $0.23 | $0.34 | 131K | −3.9% |
| OLMO-3-32B-THI AllenAI | US | $0.15 | $0.50 | 66K | −3.1% |
| PHI-4-MINI-INS Microsoft | US | $0.08 | $0.35 | 131K | +8.7% |
| LLAMA-3.3-NEMO NVIDIA | US | $0.10 | $0.40 | 131K | +2.4% |
| DEVSTRAL-2512 Mistral | EU | $0.40 | $2.00 | 262K | −13.4% |
| KIMI-K2-0905 Moonshot | CN | $0.60 | $2.50 | 262K | −14.0% |
| HERMES-4-70B Nous | US | $0.13 | $0.40 | 131K | −29.1% |
| DEEPSEEK-CHAT- DeepSeek | CN | $0.21 | $0.79 | 164K | −23.1% |
| ERNIE-4.5-VL-2 Baidu | CN | $0.14 | $0.56 | 131K | −12.4% |
| GLM-4.5-AIR Zhipu | CN | $0.13 | $0.85 | 131K | −21.0% |
| GLM-4-32B Zhipu | CN | $0.10 | $0.10 | 128K | −5.7% |
| UI-TARS-1.5-7B ByteDance | CN | $0.10 | $0.20 | 128K | +1.1% |
| HUNYUAN-A13B-I Tencent | CN | $0.14 | $0.57 | 131K | +0.9% |
| ERNIE-4.5-VL-4 Baidu | CN | $0.42 | $1.25 | 131K | −4.1% |
| MISTRAL-SMALL- Mistral | EU | $0.08 | $0.20 | 128K | +4.2% |
| NEMOTRON-NANO- NVIDIA | US | $0.04 | $0.16 | 131K | −9.0% |
| GEMMA-3N-E4B-I Google | US | $0.06 | $0.12 | 33K | −22.3% |
| SPOTLIGHT Arcee | US | $0.18 | $0.18 | 131K | −21.8% |
| VIRTUOSO-LARGE Arcee | US | $0.75 | $1.20 | 131K | −10.1% |
| REKA-FLASH-3 Reka | US | $0.10 | $0.20 | 66K | −10.7% |
| DEEPSEEK-R1-DI DeepSeek | CN | $0.29 | $0.29 | 128K | −4.8% |
| SONAR Perplexity | US | $1.00 | $1.00 | 127K | −27.3% |
| COMMAND-R7B-12 Cohere | US | $0.04 | $0.15 | 128K | +0.4% |
| DEEPSEEK-R1-05 DeepSeek | CN | $0.50 | $2.15 | 164K | +7.9% |
| NOVA-MICRO-V1 Amazon | US | $0.04 | $0.14 | 128K | −13.5% |
| NOVA-PRO-V1 Amazon | US | $0.80 | $3.20 | 300K | −7.0% |
| LLAMA-3.2-3B-I Meta | US | $0.05 | $0.34 | 131K | −13.2% |
| DEEPSEEK-V3.1- DeepSeek | CN | $0.27 | $0.95 | 164K | −17.2% |
| DEEPSEEK-R1-DI DeepSeek | CN | $0.70 | $0.80 | 131K | +7.0% |
| VOXTRAL-SMALL- Mistral | EU | $0.10 | $0.30 | 32K | −18.3% |
| GRANITE-4.0-H- IBM | US | $0.02 | $0.11 | 131K | −9.1% |
| GLM-5.1 Zhipu | CN | $0.98 | $3.08 | 203K | +6.7% |
| GLM-4.5V Zhipu | CN | $0.60 | $1.80 | 66K | −1.7% |
| CLAUDE-3.5-HAI Anthropic | US | $0.80 | $4.00 | 200K | −2.5% |
| CODER-LARGE Arcee | US | $0.50 | $0.80 | 33K | −6.6% |
| GLM-5-TURBO Zhipu | CN | $1.20 | $4.00 | 203K | −21.3% |
| CLAUDE-HAIKU-4 Anthropic | US | $1.00 | $5.00 | 200K | −13.4% |
| MAESTRO-REASON Arcee | US | $0.90 | $3.30 | 131K | −14.9% |
| GLM-5V-TURBO Zhipu | CN | $1.20 | $4.00 | 203K | −13.6% |
| MINIMAX-M2-HER MiniMax | CN | $0.30 | $1.20 | 66K | −12.9% |
| PHI-4 Microsoft | US | $0.07 | $0.14 | 16K | +7.9% |
| REKA-EDGE Reka | US | $0.10 | $0.10 | 16K | −24.2% |
| JAMBA-LARGE-1. AI21 | US | $2.00 | $8.00 | 256K | −4.2% |
| MISTRAL-MEDIUM Mistral | EU | $1.50 | $7.50 | 262K | −2.6% |
| SONAR-REASONIN Perplexity | US | $2.00 | $8.00 | 128K | −27.5% |
| SONAR-DEEP-RES Perplexity | US | $2.00 | $8.00 | 128K | −26.9% |
| COMMAND-A Cohere | US | $2.50 | $10.00 | 256K | +5.5% |
| GPT-5.3-CODEX OpenAI | US | $1.75 | $14.00 | 400K | −19.8% |
| GPT-5.2-CODEX OpenAI | US | $1.75 | $14.00 | 400K | +0.7% |
| INFLECTION-3-P Inflection | US | $2.50 | $10.00 | 8K | −16.1% |
| CLAUDE-OPUS-4. Anthropic | US | $5.00 | $25.00 | 1000K | +11.8% |
| INFLECTION-3-P Inflection | US | $2.50 | $10.00 | 8K | −22.9% |
| SONAR-PRO Perplexity | US | $3.00 | $15.00 | 200K | −7.3% |
| GPT-5.3-CHAT OpenAI | US | $1.75 | $14.00 | 128K | +9.0% |
| SONAR-PRO-SEAR Perplexity | US | $3.00 | $15.00 | 200K | −13.3% |
| GPT-5.5-PRO OpenAI | US | $30.00 | $180.00 | 1050K | −13.1% |
| GPT-5.4-PRO OpenAI | US | $30.00 | $180.00 | 1050K | −8.8% |
| GPT-5.5 OpenAI | US | $5.00 | $30.00 | 1050K | −24.3% |
Token 成本计算器
演示面向新计算经济的 AI 指数
衡量 AI 模型价格、效率、推理成本与市场动能的结构化基准。资讯会被复制,指数不会。
价格只是故事的一部分。
效率才是真相。
最便宜的模型不一定最划算。HotON.ai 以完成任务的总成本、成功率、速度、稳定性与输出质量来比较模型。
| # | 模型 | 单任务成本 | 速度 | 稳定性 | 质量 | 效率 |
|---|---|---|---|---|---|---|
| 01 | DEEPSEEK-V4-FL DeepSeek | $0.000 | 120 t/s | 100.0% | A+ | 96 |
| 02 | DEEPSEEK-V4-PR DeepSeek | $0.001 | 121 t/s | 100.0% | A+ | 96 |
| 03 | QWEN3.6-FLASH Alibaba | $0.001 | 145 t/s | 100.0% | A+ | 96 |
| 04 | NEMOTRON-3-SUP NVIDIA | $0.000 | 152 t/s | 96.5% | A+ | 96 |
| 05 | LLAMA-4-MAVERI Meta | $0.001 | 158 t/s | 100.0% | A+ | 96 |
| 06 | LLAMA-4-SCOUT Meta | $0.000 | 140 t/s | 100.0% | A+ | 96 |
| 07 | GEMINI-3.1-FLA Google | $0.001 | 172 t/s | 99.5% | A+ | 96 |
找到真正以最低成本交付最佳结果的模型。
在哪里运行 AI 更便宜?
AI 成本不只由模型价格决定,还受地区、算力供给、电力成本、延迟与可用性影响。
全球算力网络
实时区域成本信号
比较全球各地区的推理成本格局。
算力可用性
了解哪些地区的 AI 基础设施容量正变得更有吸引力。
能源感知 AI
追踪能源状况如何影响算力与推理定价。
分时成本窗口
发现某些地区在何时对 AI 负载更具成本优势。
HotON.ai 帮助市场理解 AI 成本的地理分布。
真正重要的信号
模型发布、价格变动、基础设施变化、政策更新、融资事件与市场动向——从噪音中筛选出来。
Google DeepMind Releases Gemma 4 QAT Checkpoints: Q4_0 and a New Mobile Format Cut On-Device Memory
Google will pay SpaceX $920M per month for compute
S&P 500 rejects SpaceX, also blocking entry for OpenAI and Anthropic
"We pissed off a lot of people": Giant data center plan cut 50% amid protests
Florida's lawsuit against OpenAI and CEO Altman treats ChatGPT as a defective product and public nuisance
The most interesting startups right now want to get you off your phone
This is your laptop… on AI
HotON.ai 雷达过滤市场噪音,只突出可能影响成本、访问、能力与竞争格局的关键变化。
查看全部 →视频、图像与简报——任一格式,同一信息流
HotON.ai 以视频、图像与文字交付市场情报。你想怎么看 AI 经济都行——平台兼容每一种格式。
AI 市场每周脉动
用 90 秒视频回顾本周 AI 价格、模型与基础设施的最大动向。
观看回顾全球 Token 价格热力图
一眼看清输入与输出成本的涨跌分布。
新增容量上线,推理成本回落
本周区域算力供给趋于宽松,推动推理成本指数在三个主要地区创下月度新低……
视频报告
展示片、回顾与讲解,自适应播放。
图像与信息图
图表、信息图与封面,自适应且清晰。
文字与简报
结构化文章、摘要与数据笔记。
为 AI 决策者打造的报告
关于模型定价、AI 基础设施、推理成本、市场热度与全球 AI 供应链趋势的结构化情报。
AI 市场周报
一周最重要 AI 市场变化的精炼摘要。
阅读报告→AI 月度定价报告
深入解读模型定价、Token 成本与效率趋势。
阅读报告→全球推理成本报告
AI 任务成本在不同模型与地区如何变化。
阅读报告→模型效率报告
以真实任务表现与总成本比较领先模型。
阅读报告→AI 基础设施情报
AI 经济背后的算力、能源、云与数据中心信号。
阅读报告→订阅 HotON 报告
AI 市场简报、定价报告与指数更新——直达你的邮箱。
面向开发者与机构的 AI 市场数据
通过 HotON.ai 的 API、数据流与定制情报产品,获取结构化的 AI 市场数据。
- 构建 AI 工具的开发者
- 优化 AI 成本的企业
- 追踪市场地位的模型厂商
- 关注 AI 基础设施的投资者
- 研究 AI 经济的分析师
> GET /v1/models/OPUS-4.8/price { "symbol": "OPUS-4.8", "provider": "Anthropic", "input_per_1m": 6.00, "output_per_1m": 22.50, "context_k": 500, "efficiency": 96, "availability": "99.9%", "change_24h": "+6.2%" }