AI News AI资讯 7h ago Updated 2h ago 更新于 2小时前 48

Anthropic Releases Claude Fable 5.1 and Claude Mythos 5.1: 52.6% on Terminal-Bench-Science and 75% Cheaper Cache Reads Anthropic发布Claude Fable 5.1和Claude Mythos 5.1:Terminal-Bench-Science得分52.6%,缓存读取成本降低75%

Anthropic released Claude Fable 5.1 (general availability) and Claude Mythos 5.1 (restricted), sharing the same underlying model with different safeguard layers Terminal-Bench-Science 0.1 score of 52.6% roughly doubles Fable 5's 24.7%, outperforming Opus 5 (29.0%) and GPT-5.6 Sol (22.4%) Cache read pricing drops 75% from $1.00 to $0.25 per million tokens, yielding ~25% cost reduction on typical workloads and up to 45% on agentic ones Three breaking API changes: forced tool use removed, thinking Anthropic发布Claude Fable 5.1(公开可用)与Claude Mythos 5.1(仅限审核组织),两者共享同一底层模型但安全层不同 Terminal-Bench-Science 0.1得分达52.6%,较Fable 5的24.7%近乎翻倍,显著领先Opus 5(29.0%)与GPT-5.6 Sol(22.4%) 缓存读取价格大幅下调75%($1.00→$0.25/百万token),典型负载成本降低约25%,Agentic负载最高降45% 三项API重大变更:强制工具使用已移除、思考块与模型绑定、编辑历史对话将无效化思考块 安全层优化:网络安全允许漏洞发现但禁止exploi

72
Hot 热度
65
Quality 质量
70
Impact 影响力

Analysis 深度分析

TL;DR

  • Anthropic released Claude Fable 5.1 (general availability) and Claude Mythos 5.1 (restricted), sharing the same underlying model with different safeguard layers
  • Terminal-Bench-Science 0.1 score of 52.6% roughly doubles Fable 5's 24.7%, outperforming Opus 5 (29.0%) and GPT-5.6 Sol (22.4%)
  • Cache read pricing drops 75% from $1.00 to $0.25 per million tokens, yielding ~25% cost reduction on typical workloads and up to 45% on agentic ones
  • Three breaking API changes: forced tool use removed, thinking blocks are model-bound, and editing earlier turns now invalidates thinking blocks
  • Cyber safeguards relaxed for vulnerability discovery (60% fewer interventions in Claude Code); biology safeguards fire 85% less on benign requests

Why It Matters

This release signals Anthropic's push into agentic scientific research workflows, where the 52.6% Terminal-Bench-Science score demonstrates a meaningful leap in autonomous reasoning over complex tool-using tasks. The aggressive cache pricing cut directly addresses one of the biggest cost barriers for production agentic deployments, making long-context agent loops economically viable at scale.

Technical Details

  • Architecture & Specs: Both models share identical underlying weights with 1M token context window and 128K max output tokens; adaptive thinking is always enabled
  • Benchmark Performance: Terminal-Bench 4.0 scores 55.8% (Fable) vs 60.9% (Mythos), revealing the performance cost of safeguard interventions; CursorBench 3.2.0 at 73.4%, OSWorld 2.0 at 41.7% strict
  • Pricing: Base input $10/M tokens, output $50/M tokens unchanged; cache reads dropped to $0.25/M (0.025× base vs 0.1× on other Claude models); batch processing at $5/$25 per million
  • API Breaking Changes: tool_choice set to any or tool now returns 400 (use auto with strict tool use); thinking blocks are model-bound so router/fallback chains lose reasoning when switching down; mid-conversation system/tools array edits error on accounts created after August 31, 2026
  • Safeguard & Research Impact: Mythos 5.1 designed protein binders with ~50% hit rate across 12 targets vs 10-15% baseline; Fable 5.1 generated Venus elevation maps at 2-3 km resolution; custom GPU kernels accelerated seven open-source genomics models up to 2.5×

Industry Insight

  • Agent architectures relying on conversation history manipulation (injecting reminders, rebuilding system prompts mid-turn) will need immediate refactoring; turn-scoped system messages and server-side context editing are the new patterns to adopt
  • The cache pricing shift makes long-context agentic loops significantly more economical—teams running heavy tool-use workflows should re-evaluate cost models and may see 45% reductions, potentially unlocking previously uneconomical use cases
  • The deliberate transparency around the Fable/Mythos performance gap (55.8% vs 60.9% on identical weights) sets a new industry standard for honest benchmarking; expect competitors to face pressure to disclose similar safeguard overhead figures

TL;DR

  • Anthropic发布Claude Fable 5.1(公开可用)与Claude Mythos 5.1(仅限审核组织),两者共享同一底层模型但安全层不同
  • Terminal-Bench-Science 0.1得分达52.6%,较Fable 5的24.7%近乎翻倍,显著领先Opus 5(29.0%)与GPT-5.6 Sol(22.4%)
  • 缓存读取价格大幅下调75%($1.00→$0.25/百万token),典型负载成本降低约25%,Agentic负载最高降45%
  • 三项API重大变更:强制工具使用已移除、思考块与模型绑定、编辑历史对话将无效化思考块
  • 安全层优化:网络安全允许漏洞发现但禁止exploit开发,生物安全触发率降低85%,Mythos 5.1在蛋白质结合剂设计中实现约50%命中率

为什么值得看

本文揭示了Anthropic在Agentic AI能力与成本控制上的双重突破,为开发者提供了更具性价比的科学计算与自动化工作流解决方案。同时,API架构的三项重大变更直接影响现有Agent系统的兼容性,行业需及时调整部署策略。

技术解析

  • 模型架构与规格:Fable 5.1与Mythos 5.1共享同一底层模型,区别仅在于安全层配置。两者均支持1M token上下文窗口与128K最大输出token,自适应思考功能默认开启。
  • 基准测试表现:在Terminal-Bench-Science 0.1(Agentic科学研究基准)上得分52.6%,较前代Fable 5(24.7%)提升超一倍;Terminal-Bench 4.0中Fable 5.1得55.8%,Mythos 5.1达60.9%,差距归因于安全干预成本。
  • 定价策略调整:基础输入/输出价格维持$10/$50每百万token不变,但缓存读取价格从$1.00降至$0.25(仅为其他Claude模型的1/4),批量处理价格为$5/$25。
  • API重大变更:①移除tool_choiceany/tool选项,需改用auto配合严格工具使用;②思考块与模型绑定,早期模型无法读取Fable 5.1的思考内容;③编辑历史对话(如注入/删除per-turn提醒或重建system/tools数组)将导致思考块无效。
  • 安全层优化:网络安全策略允许漏洞发现但禁止exploit开发,Claude Code中安全干预减少约60%;生物安全层在良性请求上触发率降低85%;渗透测试、exploit生成等仍重定向至Opus。

行业启示

  • Agentic AI成本结构优化:缓存读取价格大幅下调将显著降低长上下文Agent应用的运营成本,推动科学计算、代码生成等重上下文场景的规模化部署。
  • API兼容性风险:三项重大变更可能破坏依赖历史对话编辑或强制工具调用的现有Agent系统,团队需立即评估升级影响并调整架构。
  • 安全与能力的平衡:Anthropic通过分层安全策略(Fable/Mythos)在开放可用性与企业级管控间取得平衡,同时优化安全干预频率以提升实用效率,为行业提供了可借鉴的治理范式。

Disclaimer: The above content is generated by AI and is for reference only. 免责声明:以上内容由 AI 生成,仅供参考。

Claude Claude LLM 大模型 Product Launch 产品发布 Benchmark 基准测试 Evaluation 评测