Country 4/4
Provider 18/18
Cost estimator Pick a scenario or set token counts; the chart compares the per-call cost across every model in your selected currency.
Scenario
2 free model(s) excluded from the chartClick a bar to jump to its row in the table
Price landscape One dot per model. X axis: input ($/1M), Y axis: output ($/1M), log × log.
Providers:OpenAIAnthropicGooglexAIMistral AICoherePerplexityMoonshot AI (Kimi)Zhipu (Z.AI / BigModel)ByteDance Doubao (Volcano Engine)Baidu ERNIE (Qianfan)MiniMaxStepFunTencent HunyuaniFlytek Spark01.AIDeepSeekAlibaba Cloud Model Studio
Click a point to jump to its row in the table
| Batch | I/O | Capabilities | Source | |||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| OpenAI 🇺🇸 | gpt-5.4 | 1M Max out 128K | ¥16.944 | ¥101.67 | ¥1.694 | in¥8.472 out¥50.833 | → | | 2 months ago | developers.openai.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Tier price curve InputOutputCached
Notes: GPT-5.4 是上一代旗舰；建议新项目优先评估 GPT-5.5。 → https://developers.openai.com/api/docs/pricing | ||||||||||||||||||||||||||||||||||
| OpenAI 🇺🇸 | gpt-5.4-mini | 400K Max out 128K | ¥5.083 | ¥30.5 | ¥0.5083 | in¥2.542 out¥15.25 | → | | 2 months ago | developers.openai.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 中端模型；性价比高，适合大多数生产用例。 → https://developers.openai.com/api/docs/pricing | ||||||||||||||||||||||||||||||||||
| OpenAI 🇺🇸 | gpt-5.4-nano | 400K Max out 128K | ¥1.356 | ¥8.472 | ¥0.1356 | in¥0.6778 out¥4.236 | → | | 2 months ago | developers.openai.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 超低成本档；适合大批量分类、抽取、轻量摘要任务。 → https://developers.openai.com/api/docs/pricing | ||||||||||||||||||||||||||||||||||
| OpenAI 🇺🇸 | gpt-5.4-pro | 1M Max out 128K | ¥203.33 | ¥1219.99 | — | in¥101.67 out¥610 | → | | 2 months ago | developers.openai.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Tier price curve InputOutput
Notes: 上一代 Pro 推理模型；不提供缓存折扣。 → https://developers.openai.com/api/docs/pricing | ||||||||||||||||||||||||||||||||||
| OpenAI 🇺🇸 | gpt-5.5 | 1.1M Max out 128K | ¥33.889 | ¥203.33 | ¥3.389 | in¥16.944 out¥101.67 | → | | 2 months ago | developers.openai.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Tier price curve InputOutputCached
Notes: Long-context tier 在输入 token > 272K 时整次会话计 2× 输入、1.5× 输出。Standard / Batch / Flex 适用相同阶梯。 → https://developers.openai.com/api/docs/pricing | ||||||||||||||||||||||||||||||||||
| OpenAI 🇺🇸 | gpt-5.5-pro | 1.1M Max out 128K | ¥203.33 | ¥1219.99 | — | in¥101.67 out¥610 | → | | 2 months ago | developers.openai.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 面向最高质量推理/复杂任务；不提供缓存折扣。 → https://developers.openai.com/api/docs/pricing | ||||||||||||||||||||||||||||||||||
| Anthropic 🇺🇸 | claude-fable-5 | 1M Max out 128K | ¥67.777 | ¥338.89 | ¥6.778 | in¥33.889 out¥169.44 | → | | today | platform.claude.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 价位高于 Opus 档。thinking 始终开启且不返回原始思维链。要求 30 天数据保留，零数据保留（ZDR）组织无法调用。Claude Mythos 5（Project Glasswing 限定）价格与本模型相同。 → https://platform.claude.com/docs/en/about-claude/pricing | ||||||||||||||||||||||||||||||||||
| Anthropic 🇺🇸 | claude-haiku-4-5 | 200K Max out 64K | ¥6.778 | ¥33.889 | ¥0.6778 | in¥3.389 out¥16.944 | → | | 2 months ago | platform.claude.com | ||||||||||||||||||||||||
| Tiered pricing by context length
| ||||||||||||||||||||||||||||||||||
| Anthropic 🇺🇸 | claude-opus-4-5 | 200K Max out 64K | ¥33.889 | ¥169.44 | ¥3.389 | in¥16.944 out¥84.722 | → | | 2 months ago | platform.claude.com | ||||||||||||||||||||||||
| Tiered pricing by context length
| ||||||||||||||||||||||||||||||||||
| Anthropic 🇺🇸 | claude-opus-4-6 | 1M Max out 64K | ¥33.889 | ¥169.44 | ¥3.389 | in¥16.944 out¥84.722 | → | | 2 months ago | platform.claude.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: Fast Mode 不适用于本模型：请求按标准速度执行、标准价计费。 → https://platform.claude.com/docs/en/about-claude/pricing | ||||||||||||||||||||||||||||||||||
| Anthropic 🇺🇸 | claude-opus-4-7 | 1M Max out 64K | ¥33.889 | ¥169.44 | ¥3.389 | in¥16.944 out¥84.722 | → | | 2 months ago | platform.claude.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 1M 上下文窗口下统一计价。新分词器对相同文本可能多用 ~35% tokens。Fast Mode 不可用（请求会报错）。 → https://platform.claude.com/docs/en/about-claude/pricing | ||||||||||||||||||||||||||||||||||
| Anthropic 🇺🇸 | claude-opus-4-8 | 1M Max out 128K | ¥33.889 | ¥169.44 | ¥3.389 | in¥16.944 out¥84.722 | → | | today | platform.claude.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 1M 上下文窗口下统一计价。Fast Mode（research preview，仅 Claude API）输入 $10、输出 $50，不与 Batch 叠加。 → https://platform.claude.com/docs/en/about-claude/pricing | ||||||||||||||||||||||||||||||||||
| Anthropic 🇺🇸 | claude-opus-5 | 1M Max out 128K | ¥33.889 | ¥169.44 | ¥3.389 | in¥16.944 out¥84.722 | → | | today | platform.claude.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 1M 上下文窗口下统一计价。Fast Mode（research preview，仅 Claude API）输入 $10、输出 $50，不与 Batch 叠加。最小可缓存前缀 512 tokens（Opus 4.8 为 1024）。限流额度与 Opus 4.x 系列分开计算。 → https://platform.claude.com/docs/en/about-claude/pricing | ||||||||||||||||||||||||||||||||||
| Anthropic 🇺🇸 | claude-sonnet-4-5 | 200K Max out 64K | ¥20.333 | ¥101.67 | ¥2.033 | in¥10.167 out¥50.833 | → | | 2 months ago | platform.claude.com | ||||||||||||||||||||||||
| Tiered pricing by context length
| ||||||||||||||||||||||||||||||||||
| Anthropic 🇺🇸 | claude-sonnet-4-6 | 1M Max out 64K | ¥20.333 | ¥101.67 | ¥2.033 | in¥10.167 out¥50.833 | → | | 2 months ago | platform.claude.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 1M 上下文窗口下统一计价。 → https://platform.claude.com/docs/en/about-claude/pricing | ||||||||||||||||||||||||||||||||||
| Anthropic 🇺🇸 | Promo claude-sonnet-5 | 1M Max out 128K | ¥20.333 | ¥101.67 | ¥2.033 | in¥10.167 out¥50.833 | → | | today | platform.claude.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Promotion: introductory pricing: $2/$10 per MTok through 2026-08-31, then $3/$15 (until 2026-08-31)
Notes: 1M 上下文窗口下统一计价。新分词器对相同文本约多用 30% tokens（Sonnet 4.6 及更早为旧分词器）。 → https://platform.claude.com/docs/en/about-claude/pricing | ||||||||||||||||||||||||||||||||||
| Google 🇺🇸 | gemini-2.5-flash | 1M Max out 65.5K | ¥2.033 | ¥16.944 | ¥0.2033 | in¥1.017 out¥8.472 | → | | 2 months ago | ai.google.dev | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 上下文缓存存储费 $1.00/1M tokens/小时。免费层提供 RPM/RPD 配额。 → https://ai.google.dev/gemini-api/docs/pricing | ||||||||||||||||||||||||||||||||||
| Google 🇺🇸 | gemini-2.5-flash-lite | 1M Max out 65.5K | ¥0.6778 | ¥2.711 | ¥0.0678 | in¥0.3389 out¥1.356 | → | | 2 months ago | ai.google.dev | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 当前 Gemini GA 系列中最便宜的型号；适合大批量分类、抽取任务。 → https://ai.google.dev/gemini-api/docs/pricing | ||||||||||||||||||||||||||||||||||
| Google 🇺🇸 | gemini-2.5-pro | 1M Max out 65.5K | ¥8.472 | ¥67.777 | ¥0.8472 | in¥4.236 out¥33.889 | → | | 2 months ago | ai.google.dev | ||||||||||||||||||||||||
| Tiered pricing by context length
Tier price curve InputOutputCached
Notes: 上下文缓存另收存储费 $4.50/1M tokens/小时。Batch API 50% 折扣（输入与输出）。 → https://ai.google.dev/gemini-api/docs/pricing | ||||||||||||||||||||||||||||||||||
| Google 🇺🇸 | Preview gemini-3-flash-preview | 1M Max out 65.5K | ¥3.389 | ¥20.333 | ¥0.3389 | in¥1.694 out¥10.167 | → | | 2 months ago | ai.google.dev | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 免费层提供受限 RPM/RPD（按 Tier 1 文档）。上下文缓存存储费 $1.00/1M tokens/小时（未列入价格列）。Preview 状态。 → https://ai.google.dev/gemini-api/docs/pricing | ||||||||||||||||||||||||||||||||||
| Google 🇺🇸 | Preview gemini-3.1-flash-lite | 1M Max out 65.5K | ¥1.694 | ¥10.167 | ¥0.1694 | in¥0.8472 out¥5.083 | → | | 2 months ago | ai.google.dev | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 上下文缓存存储费 $1.00/1M tokens/小时。Preview 状态。 → https://ai.google.dev/gemini-api/docs/pricing | ||||||||||||||||||||||||||||||||||
| Google 🇺🇸 | Preview gemini-3.1-pro-preview | 1M Max out 65.5K | ¥13.555 | ¥81.333 | ¥1.356 | in¥6.778 out¥40.666 | → | | 2 months ago | ai.google.dev | ||||||||||||||||||||||||
| Tiered pricing by context length
Tier price curve InputOutputCached
Notes: 上下文缓存还另收存储费 $4.50/1M tokens/小时（未列入价格列）。Preview 状态。 → https://ai.google.dev/gemini-api/docs/pricing | ||||||||||||||||||||||||||||||||||
| xAI 🇺🇸 | grok-4-1-fast-non-reasoning | 2M | ¥1.356 | ¥3.389 | — | — | → | | 2 months ago | docs.x.ai | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 低成本直答变体；2M 上下文，适合大批量推理任务。 → https://docs.x.ai/docs/models | ||||||||||||||||||||||||||||||||||
| xAI 🇺🇸 | grok-4-1-fast-reasoning | 2M | ¥1.356 | ¥3.389 | — | — | → | | 2 months ago | docs.x.ai | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 低成本快推理；2M 上下文 + thinking，性价比高。 → https://docs.x.ai/docs/models | ||||||||||||||||||||||||||||||||||
| xAI 🇺🇸 | grok-4.20-0309-non-reasoning | 2M | ¥8.472 | ¥16.944 | — | — | → | | 2 months ago | docs.x.ai | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 非推理直答变体；Batch API 20-50% 折扣，Web 搜索 $5/1000 次。 → https://docs.x.ai/docs/models | ||||||||||||||||||||||||||||||||||
| xAI 🇺🇸 | grok-4.20-0309-reasoning | 2M | ¥8.472 | ¥16.944 | — | — | → | | 2 months ago | docs.x.ai | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 推理（thinking）变体；Batch API 20-50% 折扣，Web 搜索 $5/1000 次。 → https://docs.x.ai/docs/models | ||||||||||||||||||||||||||||||||||
| xAI 🇺🇸 | grok-4.20-multi-agent-0309 | 2M | ¥8.472 | ¥16.944 | — | — | → | | 2 months ago | docs.x.ai | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 多智能体（Multi-Agent）模式；Batch API 20-50% 折扣，Web 搜索 $5/1000 次。 → https://docs.x.ai/docs/models | ||||||||||||||||||||||||||||||||||
| xAI 🇺🇸 | grok-4.3 | 1M | ¥8.472 | ¥16.944 | — | — | → | | 2 months ago | docs.x.ai | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: Batch API 提供 20-50% 折扣（按规模/区域），实际批量价请以 console 为准。Web 搜索工具按 $5/1000 次单独计费。 → https://docs.x.ai/docs/models | ||||||||||||||||||||||||||||||||||
| Mistral AI 🇫🇷 | codestral | 128K | ¥2.033 | ¥6.1 | — | — | | | 2 months ago | mistral.ai | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 代码专精模型；支持 FIM（fill-in-the-middle）补全。 → https://mistral.ai/pricing | ||||||||||||||||||||||||||||||||||
| Mistral AI 🇫🇷 | magistral-small | 128K | ¥3.389 | ¥10.167 | — | — | | | 2 months ago | mistral.ai | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: Magistral 系列推理专精；同价于 Large 3 但小尺寸。 → https://mistral.ai/pricing | ||||||||||||||||||||||||||||||||||
| Mistral AI 🇫🇷 | mistral-large-3 | 256K | ¥3.389 | ¥10.167 | — | — | | | 2 months ago | mistral.ai | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: Mistral 旗舰文本模型；性价比对标 GPT-4.1 但便宜 4-5×。 → https://mistral.ai/pricing | ||||||||||||||||||||||||||||||||||
| Mistral AI 🇫🇷 | mistral-medium-3 | 128K | ¥2.711 | ¥13.555 | — | — | | | 2 months ago | mistral.ai | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 上一代 Medium；性价比仍佳。 → https://mistral.ai/pricing | ||||||||||||||||||||||||||||||||||
| Mistral AI 🇫🇷 | mistral-medium-3.5 | 256K | ¥10.167 | ¥50.833 | — | — | → | | 2 months ago | mistral.ai | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 中端旗舰；多模态 + 推理。 → https://mistral.ai/pricing | ||||||||||||||||||||||||||||||||||
| Mistral AI 🇫🇷 | mistral-small-4 | 128K | ¥1.017 | ¥4.067 | — | — | → | | 2 months ago | mistral.ai | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 低成本多模态档；适合日常对话与轻量视觉任务。 → https://mistral.ai/pricing | ||||||||||||||||||||||||||||||||||
| Cohere 🇨🇦 | command-a-03-2025 | 256K Max out 8K | ¥16.944 | ¥67.777 | — | — | | | 2 months ago | cohere.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 当前 Cohere 旗舰；强工具使用 + 高吞吐。 → https://cohere.com/pricing | ||||||||||||||||||||||||||||||||||
| Cohere 🇨🇦 | command-a-reasoning-08-2025 | 256K Max out 32K | ¥16.944 | ¥67.777 | — | — | | | 2 months ago | cohere.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 推理（thinking）变体；max output 提升至 32K。 → https://cohere.com/pricing | ||||||||||||||||||||||||||||||||||
| Cohere 🇨🇦 | command-a-vision-07-2025 | 128K Max out 8K | ¥16.944 | ¥67.777 | — | — | → | | 2 months ago | cohere.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 多模态变体；单次最多 20 张图。擅长文档/图表/OCR。 → https://cohere.com/pricing | ||||||||||||||||||||||||||||||||||
| Cohere 🇨🇦 | command-r-08-2024 | 128K Max out 4K | ¥1.017 | ¥4.067 | — | — | | | 2 months ago | cohere.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 中端模型；性价比好，适合 RAG 与一般对话。 → https://cohere.com/pricing | ||||||||||||||||||||||||||||||||||
| Cohere 🇨🇦 | command-r-plus-08-2024 | 128K Max out 4K | ¥16.944 | ¥67.777 | — | — | | | 2 months ago | cohere.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 上一代旗舰；与 Command A 同价但能力略弱。 → https://cohere.com/pricing | ||||||||||||||||||||||||||||||||||
| Cohere 🇨🇦 | command-r7b-12-2024 | 128K Max out 4K | ¥0.2542 | ¥1.017 | — | — | | | 2 months ago | cohere.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 7B 小模型；当前 Cohere 最低价档。 → https://cohere.com/pricing | ||||||||||||||||||||||||||||||||||
| Perplexity 🇺🇸 | sonar | 200K | ¥6.778 | ¥6.778 | — | — | | | 2 months ago | docs.perplexity.ai | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 搜索增强基础款；按上下文档另收 $5/8/12 per 1K requests（low/medium/high）。 → https://docs.perplexity.ai/guides/pricing | ||||||||||||||||||||||||||||||||||
| Perplexity 🇺🇸 | sonar-deep-research | 200K | ¥13.555 | ¥54.222 | — | — | | | 2 months ago | docs.perplexity.ai | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 深度研究模式；额外计费 citation tokens $2/M、reasoning tokens $3/M、search queries $5/1K。 → https://docs.perplexity.ai/guides/pricing | ||||||||||||||||||||||||||||||||||
| Perplexity 🇺🇸 | sonar-pro | 200K | ¥20.333 | ¥101.67 | — | — | | | 2 months ago | docs.perplexity.ai | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 搜索增强 Pro；另按上下文档收 $6/10/14 per 1K req；Pro Search 模式 $14-22/1K。 → https://docs.perplexity.ai/guides/pricing | ||||||||||||||||||||||||||||||||||
| Perplexity 🇺🇸 | sonar-reasoning-pro | 200K | ¥13.555 | ¥54.222 | — | — | | | 2 months ago | docs.perplexity.ai | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 搜索 + 推理 Pro；另按上下文档 $6/10/14 per 1K req。 → https://docs.perplexity.ai/guides/pricing | ||||||||||||||||||||||||||||||||||
| Moonshot AI (Kimi) 🇨🇳 | kimi-k2-thinking | 262.1K | ¥4 | ¥16 | ¥1 | — | | | 2 months ago | platform.moonshot.cn | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: K2 系列推理模型；自动上下文缓存命中按 ¥1/M 计。 → https://platform.moonshot.cn/docs/pricing/chat-k2 | ||||||||||||||||||||||||||||||||||
| Moonshot AI (Kimi) 🇨🇳 | kimi-k2-thinking-turbo | 262.1K | ¥8 | ¥58 | ¥1 | — | | | 2 months ago | platform.moonshot.cn | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: Turbo 版本：以更高输出价换取更高吞吐与并发。 → https://platform.moonshot.cn/docs/pricing/chat-k2 | ||||||||||||||||||||||||||||||||||
| Moonshot AI (Kimi) 🇨🇳 | kimi-k2.6 | 262.1K | ¥6.5 | ¥27 | ¥1.1 | — | → | | 2 months ago | platform.moonshot.cn | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: Moonshot 当前主推多模态模型；自动上下文缓存，缓存命中按 ¥1.10/M 计。 → https://platform.moonshot.cn/docs/pricing/chat-k26 | ||||||||||||||||||||||||||||||||||
| Moonshot AI (Kimi) 🇨🇳 | moonshot-v1-128k | 131.1K | ¥10 | ¥30 | — | — | | | 2 months ago | platform.moonshot.cn | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: V1 经典系列；同名 8k / 32k / 128k 三个 API id 价格不同（¥2/¥10、¥5/¥20、¥10/¥30），此处取 128k。Vision-Preview 同价。 → https://platform.moonshot.cn/docs/pricing/chat-v1 | ||||||||||||||||||||||||||||||||||
| Zhipu (Z.AI / BigModel) 🇨🇳 | glm-4.5-air | 128K | ¥0.8 | ¥6 | ¥0.16 | — | | | 2 months ago | bigmodel.cn | ||||||||||||||||||||||||
| Tiered pricing by context length
Tier price curve InputOutputCached
Notes: 中端模型；按输入长度分档计价。短输出（<200 tokens）在输入 [0,32) 区间：输出降至 2元/M（输入与缓存价不变）。缓存存储 限时免费。 → https://bigmodel.cn/pricing | ||||||||||||||||||||||||||||||||||
| Zhipu (Z.AI / BigModel) 🇨🇳 | glm-4.7 | 200K | ¥3 | ¥14 | ¥0.6 | — | | | 2 months ago | bigmodel.cn | ||||||||||||||||||||||||
| Tiered pricing by context length
Tier price curve InputOutputCached
Notes: GLM-4 系列最新；按输入长度分档计价。短输出（<200 tokens）在输入 [0,32) 区间另享折扣：输入 2元/M、输出 8元/M、缓存 0.4元/M。缓存存储 限时免费。 → https://bigmodel.cn/pricing | ||||||||||||||||||||||||||||||||||
| Zhipu (Z.AI / BigModel) 🇨🇳 | glm-4.7-flash | 200K | Free | Free | ¥0 | — | | | 2 months ago | bigmodel.cn | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 免费档；200K 上下文。 → https://bigmodel.cn/pricing | ||||||||||||||||||||||||||||||||||
| Zhipu (Z.AI / BigModel) 🇨🇳 | glm-4.7-flashx | 200K | ¥0.5 | ¥3 | ¥0.1 | — | | | 2 months ago | bigmodel.cn | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 高速低价档；200K 上下文统一计价。缓存存储 限时免费。 → https://bigmodel.cn/pricing | ||||||||||||||||||||||||||||||||||
| Zhipu (Z.AI / BigModel) 🇨🇳 | glm-5 | 200K | ¥4 | ¥18 | ¥1 | — | | | 2 months ago | bigmodel.cn | ||||||||||||||||||||||||
| Tiered pricing by context length
Tier price curve InputOutputCached
Notes: GLM-5 主流旗舰；按输入长度分档计价。缓存存储 限时免费。 → https://bigmodel.cn/pricing | ||||||||||||||||||||||||||||||||||
| Zhipu (Z.AI / BigModel) 🇨🇳 | glm-5-turbo | 200K | ¥5 | ¥22 | ¥1.2 | — | | | 2 months ago | bigmodel.cn | ||||||||||||||||||||||||
| Tiered pricing by context length
Tier price curve InputOutputCached
Notes: GLM-5 高吞吐档；定价介于 GLM-5 与 GLM-5.1 之间，按输入长度分档计价。缓存存储 限时免费。 → https://bigmodel.cn/pricing | ||||||||||||||||||||||||||||||||||
| Zhipu (Z.AI / BigModel) 🇨🇳 | glm-5.1 | 200K | ¥6 | ¥24 | ¥1.3 | — | | | 2 months ago | bigmodel.cn | ||||||||||||||||||||||||
| Tiered pricing by context length
Tier price curve InputOutputCached
Notes: 面向长程任务（Long Horizon Task）的旗舰模型；按输入长度分档计价。缓存存储 限时免费。 → https://bigmodel.cn/pricing | ||||||||||||||||||||||||||||||||||
| ByteDance Doubao (Volcano Engine) 🇨🇳 | doubao-1.5-pro-256k | 256K | ¥5 | ¥9 | — | — | | | 2 months ago | volcengine.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 1.5 Pro 长上下文档；价格固定，不分段。 → https://www.volcengine.com/docs/82379/1544106 | ||||||||||||||||||||||||||||||||||
| ByteDance Doubao (Volcano Engine) 🇨🇳 | doubao-1.5-pro-32k | 32.8K | ¥0.8 | ¥2 | ¥0.16 | — | | | 2 months ago | volcengine.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 1.5 Pro 短上下文档；缓存命中价 ¥0.16/M（输入价的 20%）。 → https://www.volcengine.com/docs/82379/1544106 | ||||||||||||||||||||||||||||||||||
| ByteDance Doubao (Volcano Engine) 🇨🇳 | doubao-seed-1.6 | 256K | ¥0.8 | ¥8 | — | — | → | | 2 months ago | volcengine.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Tier price curve InputOutput
Notes: All-in-one 1.6 旗舰；支持深度思考、多模态、GUI 操作。特惠区（输入≤32K、输出≤200 tokens）输出价进一步降至 ¥2/M。 → https://www.volcengine.com/docs/82379/1544106 | ||||||||||||||||||||||||||||||||||
| ByteDance Doubao (Volcano Engine) 🇨🇳 | Free quota doubao-seed-1.6-flash | 256K | ¥0.15 | ¥1.5 | — | — | → | | 2 months ago | volcengine.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Tier price curve InputOutput
Notes: 极速版本：TOPT 仅 ~10ms。短上下文档极便宜（¥0.15/M），长上下文档与 Seed 1.6 同价。 → https://www.volcengine.com/docs/82379/1593704 | ||||||||||||||||||||||||||||||||||
| ByteDance Doubao (Volcano Engine) 🇨🇳 | doubao-seed-1.6-thinking | 256K | ¥0.8 | ¥8 | — | — | → | | 2 months ago | volcengine.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Tier price curve InputOutput
Notes: 强化深度推理能力的变体；阶梯定价与 Seed 1.6 一致。 → https://www.volcengine.com/docs/82379/1544106 | ||||||||||||||||||||||||||||||||||
| Baidu ERNIE (Qianfan) 🇨🇳 | ernie-4.5 | 128K | ¥4 | ¥16 | — | — | → | | 2 months ago | cloud.baidu.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 初版 4.5；建议新项目优先评估 ernie-4.5-turbo（同代降价后）。 → https://cloud.baidu.com/doc/WENXINWORKSHOP/s/hlrk4saadq | ||||||||||||||||||||||||||||||||||
| Baidu ERNIE (Qianfan) 🇨🇳 | ernie-4.5-turbo | 128K | ¥0.8 | ¥3.2 | — | — | → | | 2 months ago | cloud.baidu.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 相比初版 ERNIE 4.5 降价约 80%。多模态旗舰，性价比高。 → https://cloud.baidu.com/doc/WENXINWORKSHOP/s/hlrk4saadq | ||||||||||||||||||||||||||||||||||
| Baidu ERNIE (Qianfan) 🇨🇳 | ernie-x1 | 128K | ¥2 | ¥8 | — | — | → | | 2 months ago | cloud.baidu.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: X1 系列推理模型；推理质量更高，价格高于 Turbo。 → https://cloud.baidu.com/doc/WENXINWORKSHOP/s/hlrk4saadq | ||||||||||||||||||||||||||||||||||
| Baidu ERNIE (Qianfan) 🇨🇳 | ernie-x1-turbo | 128K | ¥1 | ¥4 | — | — | → | | 2 months ago | cloud.baidu.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: X1 系列推理模型 Turbo 档；号称推理成本约 DeepSeek R1 的 25%。 → https://cloud.baidu.com/doc/WENXINWORKSHOP/s/hlrk4saadq | ||||||||||||||||||||||||||||||||||
| MiniMax 🇨🇳 | minimax-m1 | 1M | ¥2.711 | ¥11.929 | — | — | | | 2 months ago | platform.minimax.io | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: M 系列首代；以超长 1M 上下文与高效推理为特点。 → https://platform.minimax.io/docs/guides/pricing-paygo | ||||||||||||||||||||||||||||||||||
| MiniMax 🇨🇳 | minimax-m2 | 204.8K Max out 131.1K | ¥2.033 | ¥8.133 | — | — | | | 2 months ago | platform.minimax.io | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 上一代旗舰；CNY 价 ¥2.1/¥8.4。 → https://platform.minimax.io/docs/guides/pricing-paygo | ||||||||||||||||||||||||||||||||||
| MiniMax 🇨🇳 | minimax-m2.5 | 204.8K Max out 131.1K | ¥1.017 | ¥7.794 | — | — | | | 2 months ago | platform.minimax.io | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 当前 MiniMax 旗舰；230B 总参 / 10B 激活 MoE。国内站 platform.minimaxi.com 可能以 CNY 单独计价（参考 ¥1/¥8）。 → https://platform.minimax.io/docs/guides/pricing-paygo | ||||||||||||||||||||||||||||||||||
| StepFun 🇨🇳 | step-1v-8k | 8K | ¥5 | ¥20 | — | — | → | | 2 months ago | platform.stepfun.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: V1 多模态版本；8K 上下文，适合短输入图文理解。 → https://platform.stepfun.com/docs/zh/pricing/details | ||||||||||||||||||||||||||||||||||
| StepFun 🇨🇳 | step-2-mini | 16.4K | ¥1 | ¥2 | — | — | | | 2 months ago | platform.stepfun.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: Step 2 系列高性价比版；适合大批量轻量任务。 → https://platform.stepfun.com/docs/zh/pricing/details | ||||||||||||||||||||||||||||||||||
| StepFun 🇨🇳 | step-3 | 256K | ¥1.5 | ¥4 | — | — | → | | 2 months ago | platform.stepfun.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 321B 总参 / 38B 激活 MoE；强视觉感知 + 复杂推理；开源权重可下载。当前为促销价。 → https://platform.stepfun.com/docs/zh/pricing/details | ||||||||||||||||||||||||||||||||||
| Tencent Hunyuan 🇨🇳 | hunyuan-a13b | 128K | ¥0.5 | ¥2 | — | — | | | 2 months ago | cloud.tencent.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 低成本 13B 激活档；适合大批量轻量任务。 → https://cloud.tencent.com/document/product/1729/97731 | ||||||||||||||||||||||||||||||||||
| Tencent Hunyuan 🇨🇳 | hunyuan-t1 | 128K | ¥1 | ¥4 | — | — | | | 2 months ago | cloud.tencent.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: T 系列推理（thinking）模型。 → https://cloud.tencent.com/document/product/1729/97731 | ||||||||||||||||||||||||||||||||||
| Tencent Hunyuan 🇨🇳 | Free quota hunyuan-turbo-s | 256K | ¥0.8 | ¥2 | — | — | | | 2 months ago | cloud.tencent.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 当前主推快推理；与混元生文/多模态系列共享免费额度。 → https://cloud.tencent.com/document/product/1729/97731 | ||||||||||||||||||||||||||||||||||
| Tencent Hunyuan 🇨🇳 | hy-2.0-think | 32K | ¥3.975 | ¥15.9 | — | — | | | 2 months ago | cloud.tencent.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: HY 2.0 思考版（高端档）。自 2026-02-06 起暂不支持预付费资源包。 → https://cloud.tencent.com/document/product/1729/97731 | ||||||||||||||||||||||||||||||||||
| iFlytek Spark 🇨🇳 | spark-lite | 8K | Free | Free | — | — | | | 2 months ago | xinghuo.xfyun.cn | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 科大讯飞**永久免费**开放；适合低算力推理与联网搜索场景。 → https://xinghuo.xfyun.cn/sparkapi | ||||||||||||||||||||||||||||||||||
| iFlytek Spark 🇨🇳 | spark-max-32k | 32K | ¥21 | ¥21 | — | — | | | 2 months ago | xinghuo.xfyun.cn | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: API 名 max-32k；公开报道价 ¥0.21/万 tokens；输入与输出价格未拆分公示，此处按统一档位记 ¥21/M。 → https://xinghuo.xfyun.cn/sparkapi | ||||||||||||||||||||||||||||||||||
| iFlytek Spark 🇨🇳 | spark-pro-128k | 128K | ¥21 | ¥21 | — | — | | | 2 months ago | xinghuo.xfyun.cn | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: API 名 pro-128k；长上下文版本，价格与 Max 同档（¥21/M 估算）。 → https://xinghuo.xfyun.cn/sparkapi | ||||||||||||||||||||||||||||||||||
| 01.AI 🇨🇳 | yi-large | 32K | ¥13.555 | ¥13.555 | — | — | | | 2 months ago | platform.lingyiwanwu.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 上一代旗舰；输入与输出统一价。 → https://platform.lingyiwanwu.com/docs | ||||||||||||||||||||||||||||||||||
| 01.AI 🇨🇳 | yi-lightning | 16K | ¥0.9489 | ¥0.9489 | — | — | | | 2 months ago | platform.lingyiwanwu.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 当前 01.AI 旗舰；输入与输出统一价 ≈ ¥0.99/M。LMSYS 排行榜国内首破 GPT-4o。 → https://platform.lingyiwanwu.com/docs | ||||||||||||||||||||||||||||||||||
| DeepSeek 🇨🇳 | deepseek-v4-flash | 1M Max out 384K | ¥1 | ¥2 | ¥0.02 | — | | | 3 months ago | api-docs.deepseek.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: deepseek-chat（非 thinking 模式入口）映射至此模型；FIM 仅在非 thinking 模式可用。 → https://api-docs.deepseek.com/quick_start/pricing | ||||||||||||||||||||||||||||||||||
| DeepSeek 🇨🇳 | Promo deepseek-v4-pro | 1M Max out 384K | ¥3 | ¥6 | ¥0.025 | — | | | 3 months ago | api-docs.deepseek.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Promotion: 2.5 折优惠（页面所示已为优惠后价格） (until 2026-05-31)
Notes: deepseek-reasoner（thinking 模式入口）映射至此模型；FIM 仅在非 thinking 模式可用。 → https://api-docs.deepseek.com/quick_start/pricing | ||||||||||||||||||||||||||||||||||
| Alibaba Cloud Model Studio 🇨🇳 | Free quota qwen-max | 32.8K | ¥10.844 | ¥43.377 | — | in¥5.422 out¥21.689 | | | 3 months ago | alibabacloud.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Notes: 不分上下文阶梯。批量调用：输入与输出均 50% 折扣。 → https://www.alibabacloud.com/help/en/model-studio/model-pricing | ||||||||||||||||||||||||||||||||||
| Alibaba Cloud Model Studio 🇨🇳 | Free quota qwen3-max | 252K | ¥8.133 | ¥40.666 | — | in¥4.067 out¥20.333 | | | 3 months ago | alibabacloud.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Tier price curve InputOutput
Notes: 上下文缓存命中：输入价 50% 折扣（按所在阶梯计算）。批量调用：输入与输出均 50% 折扣。 → https://www.alibabacloud.com/help/en/model-studio/model-pricing | ||||||||||||||||||||||||||||||||||
| Alibaba Cloud Model Studio 🇨🇳 | Preview Free quota qwen3-max-preview | 252K | ¥8.133 | ¥40.666 | — | — | | | 3 months ago | alibabacloud.com | ||||||||||||||||||||||||
| Tiered pricing by context length
Tier price curve InputOutput
Notes: 预览版本不支持 Batch API；上下文缓存命中：输入价 50% 折扣。 → https://www.alibabacloud.com/help/en/model-studio/model-pricing | ||||||||||||||||||||||||||||||||||
No matching models