PulseAugur
中
实时 14:13:17
English(EN) AI API price comparison 2026: DeepSeek V4 Flash — .014/1M input GPT-4.1-mini — .40/1M Gemini 2.5 Flash — .15/1M Claude Haiku 4.5 — .80/1M That is 28x cheaper. A

Gemini-3.5-flash 在俄语文本上媲美 GPT-5.5；中国模型以更低价格击败竞争对手

作者 PulseAugur 编辑部 · [9 个来源] ·

新的基准测试显示，谷歌的 Gemini-3.5-flash 在长篇俄语内容上与 OpenAI 的 GPT-5.5 相当，成本降低了 2.5 倍。中国模型也展示了显著的价格性能优势，DeepSeek V4 Pro 在降价后更具成本效益。腾讯的 Hy3-preview 为 A 级性能提供了特别低成本的选择。多个来源指出 DeepSeek V4 Flash 比 GPT-4.1-mini 便宜多达 28 倍，可通过 Token China 提供的兼容 OpenAI 的 API 访问。 AI

影响 加速了经济高效的 LLM 的采用，迫使 OpenAI 和 Anthropic 等现有公司在价格上竞争。

排序理由 多个来源报道了各种 LLM 的新性能基准和价格比较，突出了某些模型显著的成本优势。

在 Mastodon — fosstodon.org 阅读 →

AI 生成摘要 · Google Gemini · 来自 9 个来源。 我们如何撰写摘要 →

报道来源 [9]

  1. Mastodon — fosstodon.org TIER_1 Русский(RU) · [email protected] ·

    Gemini-3.5-flash caught up with GPT-5.5 at 97/S and is 2.5x cheaper. But the main thing is that the Chinese are winning in price and quality A month ago I wrote about the DeepSeek V4 Pro paradox —

    Gemini-3.5-flash догнал GPT-5.5 на 97/S и в 2.5× дешевле. Но главное — китайцы выигрывают по цене и качеству Месяц назад я писал про парадокс DeepSeek V4 Pro — модель проиграла собственному Flash и Qwen 3.6 Plus трёхнедельной давности. Сегодня прогнал свежий battle на пяти моделя…

    链接 habr.com/…/1040770
  2. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] ·

    Token China API pricing: deepseek-v4-flash at 0.1x rate. That is 10x lower than baseline. glm-5.1 at 0.5x. glm-5v-turbo at 0.72x. deepseek-v4-pro at 1.75x for h

    Token China API pricing: deepseek-v4-flash at 0.1x rate. That is 10x lower than baseline. glm-5.1 at 0.5x. glm-5v-turbo at 0.72x. deepseek-v4-pro at 1.75x for heavy reasoning. One key, 4 models. OpenAI-compatible. No phone. USDT. https:// token-china.cc # AI # API

    链接 token-china.cc
  3. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] ·

    DeepSeek V4 Flash: .014/1M. GPT-4.1-mini: .40/1M. Claude Haiku 4.5: .80/1M. One of these is 28x cheaper. One line of code to switch. https:// token-china.cc # A

    DeepSeek V4 Flash: .014/1M. GPT-4.1-mini: .40/1M. Claude Haiku 4.5: .80/1M. One of these is 28x cheaper. One line of code to switch. https:// token-china.cc # AI # pricing # DeepSeek # OpenAI

    链接 token-china.cc
  4. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] ·

    AI API cost reality check: /month with GPT-4.1-mini vs /month with DeepSeek V4 Flash. Same work. 90% cheaper. One line change: base_url = https:// token-china.c

    AI API cost reality check: /month with GPT-4.1-mini vs /month with DeepSeek V4 Flash. Same work. 90% cheaper. One line change: base_url = https:// token-china.cc/v1 # AI # buildinpublic # SaaS # pricing

  5. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] ·

    Why are you still paying GPT-4.1-mini .40 per 1M tokens when DeepSeek V4 Flash is .014? Same quality. Same OpenAI-compatible API. 28x cheaper. No phone needed.

    Why are you still paying GPT-4.1-mini .40 per 1M tokens when DeepSeek V4 Flash is .014? Same quality. Same OpenAI-compatible API. 28x cheaper. No phone needed. No KYC. https:// token-china.cc # AI # GPT # pricing

    链接 token-china.cc
  6. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] ·

    DeepSeek V4 Flash: .014/1M tokens. GPT-4.1-mini: .40/1M tokens. That is not a typo. 28x cheaper. One base_url change. Same OpenAI SDK code. No phone verificatio

    DeepSeek V4 Flash: .014/1M tokens. GPT-4.1-mini: .40/1M tokens. That is not a typo. 28x cheaper. One base_url change. Same OpenAI SDK code. No phone verification. https:// token-china.cc # AI # coding # pricing

    链接 token-china.cc
  7. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] ·

    28x cheaper than GPT-4.1-mini. Same code. Same quality. DeepSeek V4 Flash via Token China: .014/1M input. OpenAI: .40/1M. You are burning money if you are not u

    28x cheaper than GPT-4.1-mini. Same code. Same quality. DeepSeek V4 Flash via Token China: .014/1M input. OpenAI: .40/1M. You are burning money if you are not using this. https:// token-china.cc # AI # API # DeepSeek # GPT4

    链接 token-china.cc
  8. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] ·

    DeepSeek V4 Flash vs GPT-4.1-mini pricing: ~.014 vs .40 per 1M input tokens. Same tasks, fraction of the cost. Access both DeepSeek and GLM through Token China.

    DeepSeek V4 Flash vs GPT-4.1-mini pricing: ~.014 vs .40 per 1M input tokens. Same tasks, fraction of the cost. Access both DeepSeek and GLM through Token China. One API key. https:// token-china.cc # AI

    链接 token-china.cc
  9. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] ·

    AI API price comparison 2026: DeepSeek V4 Flash — .014/1M input GPT-4.1-mini — .40/1M Gemini 2.5 Flash — .15/1M Claude Haiku 4.5 — .80/1M That is 28x cheaper. A

    AI API price comparison 2026: DeepSeek V4 Flash — .014/1M input GPT-4.1-mini — .40/1M Gemini 2.5 Flash — .15/1M Claude Haiku 4.5 — .80/1M That is 28x cheaper. And no phone verification. https:// token-china.cc # AI # API # pricing

    链接 token-china.cc

相关实体

相关话题