附件 · PDF p.29 · Table 2.16.A · Claude Sonnet 4.6 subject column
材料版本
PDF with changelog dated 2026-03-06; 135 pages; collected 2026-11-10
版本公布
未确认
披露时间
未确认 · 首发/后补未确认
采集时间
历史依据
暂无可靠历史版本依据
引用文本
Claude Sonnet 4.6 · OpenAI MRCR(v2, 1M context, 8 needles),64K extended thinking配置,Mean Match Ratio 65.1(具体成绩公布时间未确认);Claude Sonnet 4.6 · OpenAI MRCR(v2, 1M context, 8 needles),Adaptive thinking, max effort配置,Mean Match Ratio 65.8(具体成绩公布时间未确认)。官方出处 Claude Sonnet 4.6 System Card(附件,PDF with changelog dated 2026-03-06; 135 pages; collected 2026-11-10)PDF p.29 · Table 2.16.A · Claude Sonnet 4.6 subject column,https://www.anthropic.com/claude-sonnet-4-6-system-card#page=29。Benchlog https://benchlog.lekshome.cc/evidence/claude-sonnet-4-6/ev46-mrcr-1m/
关联成绩与条件
Claude Sonnet 4.6 · OpenAI MRCR 成绩与配置
指标与成绩
配置与评测条件
证据关系
Mean Match Ratio65.1原文 65.1 (64k) · 单位 未标注
公布 未确认
64K extended thinking
设置
Extended thinking with a 64K token budget.
未知项
Effort setting is not stated for the long-context table results.
工具
工具条件未记录
推理预算
思考预算 64K
框架
未记录
协议
OpenAI MRCR v2, 1M context bin, 8 needles; full internal problem set.
条件
Five-trial average; 1M context window; default sampling. Some prompts exceed public API's 1M token limit, so the full-set result is not reproducible through that API.
直接支持该成绩
Mean Match Ratio65.8原文 65.8 (max) · 单位 未标注
公布 未确认
Adaptive thinking, max effort
设置
Adaptive thinking enabled; max effort; default temperature and top_p unless an evaluation specifies otherwise.
未知项
Context window is evaluation-dependent; Table 2.1.A states it does not exceed 1M tokens.
工具
工具条件未记录
推理预算
max
框架
未记录
协议
OpenAI MRCR v2, 1M context bin, 8 needles; full internal problem set.
条件
Five-trial average; 1M context window; default sampling. Some prompts exceed public API's 1M token limit, so the full-set result is not reproducible through that API.
Official source for OpenAI MRCR · v2, 1M context, 8 needles in Claude Sonnet 4.6. See “Claude Sonnet 4.6 System Card” at PDF p.29 · Table 2.16.A · Claude Sonnet 4.6 subject column. Evidence ev46-mrcr-1m.
Claude Sonnet 4.6 · OpenAI MRCR (v2, 1M context, 8 needles), configuration 64K extended thinking, Mean Match Ratio 65.1 (exact score publication date unconfirmed). See “Claude Sonnet 4.6 System Card” at PDF p.29 · Table 2.16.A · Claude Sonnet 4.6 subject column. View score conditions
Claude Sonnet 4.6 · OpenAI MRCR (v2, 1M context, 8 needles), configuration Adaptive thinking, max effort, Mean Match Ratio 65.8 (exact score publication date unconfirmed). See “Claude Sonnet 4.6 System Card” at PDF p.29 · Table 2.16.A · Claude Sonnet 4.6 subject column. View score conditions
OpenAI MRCR v2 1M 8-needles (Mean Match Ratio) | Claude Sonnet 4.6 | 65.1 (64k) | 65.8 (max)
Attachment · PDF p.29 · Table 2.16.A · Claude Sonnet 4.6 subject column
Material version
PDF with changelog dated 2026-03-06; 135 pages; collected 2026-11-10
Version published
Unconfirmed
Disclosure date
Unconfirmed · Launch or later disclosure unconfirmed
Collected
Historical basis
No reliable historical version evidence available
Citation text
Claude Sonnet 4.6 · OpenAI MRCR (v2, 1M context, 8 needles), configuration 64K extended thinking, Mean Match Ratio 65.1 (exact score publication date unconfirmed); Claude Sonnet 4.6 · OpenAI MRCR (v2, 1M context, 8 needles), configuration Adaptive thinking, max effort, Mean Match Ratio 65.8 (exact score publication date unconfirmed). Official source: Claude Sonnet 4.6 System Card (Attachment, PDF with changelog dated 2026-03-06; 135 pages; collected 2026-11-10), PDF p.29 · Table 2.16.A · Claude Sonnet 4.6 subject column, https://www.anthropic.com/claude-sonnet-4-6-system-card#page=29. Benchlog https://benchlog.lekshome.cc/evidence/claude-sonnet-4-6/ev46-mrcr-1m/?lang=en
Related scores and conditions
Claude Sonnet 4.6 · OpenAI MRCR scores and configurations
Metric and score
Configuration and evaluation conditions
Evidence relation
Mean Match Ratio65.1Reported 65.1 (64k) · Unit not specified
Published Unconfirmed
64K extended thinking
Settings
Extended thinking with a 64K token budget.
Unknowns
Effort setting is not stated for the long-context table results.
Tools
Tool access not recorded
Reasoning budget
Thinking budget 64K
Framework
not recorded
Protocol
OpenAI MRCR v2, 1M context bin, 8 needles; full internal problem set.
Conditions
Five-trial average; 1M context window; default sampling. Some prompts exceed public API's 1M token limit, so the full-set result is not reproducible through that API.
Directly supports this score
Mean Match Ratio65.8Reported 65.8 (max) · Unit not specified
Published Unconfirmed
Adaptive thinking, max effort
Settings
Adaptive thinking enabled; max effort; default temperature and top_p unless an evaluation specifies otherwise.
Unknowns
Context window is evaluation-dependent; Table 2.1.A states it does not exceed 1M tokens.
Tools
Tool access not recorded
Reasoning budget
max
Framework
not recorded
Protocol
OpenAI MRCR v2, 1M context bin, 8 needles; full internal problem set.
Conditions
Five-trial average; 1M context window; default sampling. Some prompts exceed public API's 1M token limit, so the full-set result is not reproducible through that API.