跳至主要内容
OHLAC AI
说明中心 法律与隐私 返回首页

本页目录

  1. 模型总览
  2. 怎么选模型

On this page

  1. Model overview
  2. How to choose

このページの目次

  1. Model overview
  2. How to choose

En esta página

  1. Model overview
  2. How to choose

Help & Docs

模型说明

按旗舰 / 次旗舰 / 高性价比分组,了解定位、能力与选型建议。

版本
0.3.0
生效日期
2026 年 8 月 10 日
最后更新
2026 年 9 月 1 日

了解 OHLAC 默幕 中各模型的定位、能力与适合场景。模型按产品 Badge 分组,便于快速判断「用来干什么、大概消耗多少」。

积分消耗以很低 / 低 / 中 / 高 / 很高作相对参考,实际结算以服务端真实用量为准。了解积分规则 →

模型总览

旗舰模型

Claude Fable 5 旗舰

旗舰创作与分析,适合复杂表达、长文处理和高要求写作。

推荐场景
顶级创作、深度分析、复杂写作
速度
慢
能力等级
旗舰
积分消耗
很高
图片能力
暂未开放
产品标签
创意写作复杂分析长文处理
GPT-5.6 Sol 旗舰

最强推理模型,适合重要决策、工作规划、复杂分析。

推荐场景
重要决策、复杂分析、工作规划
速度
较慢
能力等级
深度
积分消耗
高
图片能力
图片理解
产品标签
复杂分析工作建议
Claude Opus 4.8 旗舰

强项在复杂问题,适合工作规划、复杂分析和长文处理。

推荐场景
复杂分析、长文处理、方案梳理
速度
较慢
能力等级
深度
积分消耗
高
图片能力
图片理解
产品标签
复杂分析工作建议长文处理
Grok 4.6 旗舰

旗舰级开放思考,适合重要决策、复杂分析和创意写作。

推荐场景
重要决策、开放讨论、复杂分析
速度
中
能力等级
均衡偏强
积分消耗
低
图片能力
图片理解
产品标签
复杂分析创意写作工作建议

次旗舰模型

GPT-5.6 Terra 次旗舰

综合能力强,适合日常交流、工作协助和代码开发。

推荐场景
日常对话、写作、代码开发
速度
中
能力等级
均衡
积分消耗
中
图片能力
图片理解
产品标签
工作建议编程助手日常问答
Claude Opus 4.7 次旗舰

擅长把复杂内容讲清楚,适合学习辅导、长文处理和专业写作。

推荐场景
学习辅导、长文处理、专业写作
速度
较慢
能力等级
深度
积分消耗
高
图片能力
图片理解
产品标签
学习辅导创意写作长文处理
Claude Opus 4.6 次旗舰

稳定处理难题,适合方案比较、资料整理和专业问答。

推荐场景
方案比较、资料整理、专业问答
速度
较慢
能力等级
深度
积分消耗
高
图片能力
图片理解
产品标签
工作建议复杂分析学习辅导
Claude Sonnet 5 次旗舰

综合能力强,适合工作协助、日常交流和代码开发。

推荐场景
工作协助、日常交流、代码开发
速度
中
能力等级
均衡偏强
积分消耗
中
图片能力
图片理解
产品标签
工作建议编程助手日常问答

高性价比模型

GPT-5.6 Luna 高性价比

轻快省心,适合日常问答、信息整理和生活咨询。

推荐场景
快速问答、信息整理、生活咨询
速度
快
能力等级
轻量
积分消耗
低
图片能力
图片理解
产品标签
日常问答生活咨询
Gemini 3.5 Flash 高性价比

轻快高效,适合日常问答、学习辅导和资料整理。

推荐场景
日常问答、学习辅导、资料整理
速度
快
能力等级
轻量
积分消耗
低
图片能力
图片理解
产品标签
日常问答学习辅导长文处理
Gemini 3.7 Flash 高性价比

轻快灵活,适合日常问答、头脑风暴和日常整理。

推荐场景
日常问答、头脑风暴、日常整理
速度
快
能力等级
轻量
积分消耗
低
图片能力
图片理解
产品标签
日常问答创意写作工作建议
Grok 4.5 高性价比

均衡而有想法,适合开放对话、创意写作和工作讨论。

推荐场景
开放对话、创意写作、工作讨论
速度
中
能力等级
均衡
积分消耗
低
图片能力
图片理解
产品标签
创意写作工作建议感情交流

其他模型

OHLAC 1.0

轻量可靠,适合日常交流、生活咨询和简单改写。

推荐场景
日常交流、生活咨询、简单改写
速度
很快
能力等级
轻量
积分消耗
低
图片能力
图片理解
产品标签
生活咨询日常问答
OHLAC 1.5

综合能力均衡,适合工作协助、学习辅导和日常写作。

推荐场景
工作协助、学习辅导、日常写作
速度
快
能力等级
均衡
积分消耗
低
图片能力
图片理解
产品标签
工作建议学习辅导创意写作
OHLAC Pro

更擅长拆解难题,适合重要决定、方案梳理和长文处理。

推荐场景
重要决定、方案梳理、长文处理、文档阅读
速度
中
能力等级
深度
积分消耗
中
图片能力
图片理解(文档/PDF 文字提取)
文件能力
PDF、文档(OCR 文字提取)
产品标签
复杂分析工作建议长文处理
GPT-5.5 Instant

回复迅速,适合日常问答、文字整理和轻量沟通。

推荐场景
即时问答、文字整理、日常沟通
速度
快
能力等级
均衡偏深
积分消耗
高
图片能力
图片理解
产品标签
日常问答工作建议生活咨询
Claude Sonnet 4.6

稳健通用,适合工作协助、学习辅导和代码开发。

推荐场景
工作协助、学习辅导、代码开发
速度
中
能力等级
均衡
积分消耗
中
图片能力
暂未开放
产品标签
工作建议学习辅导编程助手
Gemini 3.1 Pro Preview

善于处理复杂资料,适合学习辅导、长文处理和复杂分析。

推荐场景
长文档分析、学习辅导、复杂分析
速度
中
能力等级
深度
积分消耗
中
图片能力
图片理解
产品标签
学习辅导长文处理复杂分析

怎么选模型

根据任务类型快速匹配(建议性):

普通用户 / 日常使用

GPT-5.6 LunaGemini 3.5 FlashGemini 3.7 FlashGrok 4.5OHLAC 1.0OHLAC 1.5

复杂问题 / 重要决策

Claude Fable 5GPT-5.6 SolClaude Opus 4.8Grok 4.6

写作与表达

OHLAC 1.5GPT-5.6 TerraClaude Sonnet 5Grok 4.5Grok 4.6

代码开发

GPT-5.6 TerraClaude Sonnet 5Claude Opus 4.8

学习辅导 / 资料整理

Gemini 3.1 Pro PreviewClaude Opus 4.7Gemini 3.5 Flash

图片理解

多数模型已支持OHLAC Pro(文档/PDF OCR)

开放对话 / 头脑风暴

Grok 4.5Grok 4.6

选型原则

先选够用且更省积分的模型,只有任务确实需要更强能力时再升级。了解积分规则 →

说明

  • 速度:相对速度,便于选择,不代表性能承诺(很快 > 快 > 中 > 较慢 > 慢)。
  • 能力等级:轻量 < 均衡 < 均衡偏深/偏强 < 深度 < 旗舰。
  • 积分消耗级别:相对参考;真实结算以服务端用量为准。
  • 图片 / 文件能力:以产品内实际支持为准;未标注开放的能力不会向用户宣传。
  • OHLAC Auto:规划中,尚未上线,请勿理解为已自动选模。

Help & Docs

Models

Grouped by flagship / near-flagship / great value — positioning, capabilities, and how to choose.

Version
0.3.0
Effective date
2026-08-10
Last updated
2026-09-01

Explore OHLAC 默幕 models by product badge — so you can quickly see what each model is for and how much it tends to cost.

Credit cost is shown as very low / low / medium / high / very high for relative guidance only. Final settlement follows actual server-side usage. Credit rules →

Model overview

Flagship

Claude Fable 5 Flagship

Flagship creation and analysis — complex expression, long documents, and high-bar writing.

Best for
Top-tier creation, deep analysis, complex writing
Speed
Slow
Capability
Flagship
Credits
Very high
Image
Not available
Tags
Creative writingcomplex analysislong documents
GPT-5.6 Sol Flagship

Strongest reasoning model — important decisions, planning, and complex analysis.

Best for
Important decisions, complex analysis, work planning
Speed
Slower
Capability
Deep
Credits
High
Image
Image understanding
Tags
Complex analysiswork advice
Claude Opus 4.8 Flagship

Excels at hard problems — planning, complex analysis, and long documents.

Best for
Complex analysis, long documents, structuring plans
Speed
Slower
Capability
Deep
Credits
High
Image
Image understanding
Tags
Complex analysiswork advicelong documents
Grok 4.6 Flagship

Flagship open thinking — important decisions, complex analysis, and creative writing.

Best for
Important decisions, open discussion, complex analysis
Speed
Medium
Capability
Balanced+
Credits
Low
Image
Image understanding
Tags
Complex analysiscreative writingwork advice

Near-flagship

GPT-5.6 Terra Near-flagship

Strong all-rounder — daily chat, work help, and coding.

Best for
Daily chat, writing, coding
Speed
Medium
Capability
Balanced
Credits
Medium
Image
Image understanding
Tags
Work advicecodingdaily Q&A
Claude Opus 4.7 Near-flagship

Makes complex content clear — tutoring, long documents, and professional writing.

Best for
Tutoring, long documents, professional writing
Speed
Slower
Capability
Deep
Credits
High
Image
Image understanding
Tags
Learningcreative writinglong documents
Claude Opus 4.6 Near-flagship

Steady on hard problems — comparing options, organizing material, and expert Q&A.

Best for
Comparing options, organizing material, expert Q&A
Speed
Slower
Capability
Deep
Credits
High
Image
Image understanding
Tags
Work advicecomplex analysistutoring
Claude Sonnet 5 Near-flagship

Strong all-rounder — work help, daily chat, and coding.

Best for
Work help, daily chat, coding
Speed
Medium
Capability
Balanced+
Credits
Medium
Image
Image understanding
Tags
Work advicecodingdaily Q&A

Great value

GPT-5.6 Luna Great value

Light and easy — daily Q&A, organizing info, and life questions.

Best for
Quick Q&A, organizing info, life questions
Speed
Fast
Capability
Lightweight
Credits
Low
Image
Image understanding
Tags
Daily Q&Alife advice
Gemini 3.5 Flash Great value

Fast and efficient — daily Q&A, tutoring, and organizing material.

Best for
Daily Q&A, tutoring, organizing material
Speed
Fast
Capability
Lightweight
Credits
Low
Image
Image understanding
Tags
Daily Q&Atutoringlong documents
Gemini 3.7 Flash Great value

Fast and flexible — daily Q&A, brainstorming, and everyday organizing.

Best for
Daily Q&A, brainstorming, everyday organizing
Speed
Fast
Capability
Lightweight
Credits
Low
Image
Image understanding
Tags
Daily Q&Acreative writingwork advice
Grok 4.5 Great value

Balanced with personality — open dialogue, creative writing, and work discussion.

Best for
Open dialogue, creative writing, work discussion
Speed
Medium
Capability
Balanced
Credits
Low
Image
Image understanding
Tags
Creative writingwork adviceconversation

Other models

OHLAC 1.0

Light and reliable — daily chat, life questions, and simple rewriting.

Best for
Daily chat, life questions, simple rewriting
Speed
Very fast
Capability
Lightweight
Credits
Low
Image
Image understanding
Tags
Life advicedaily Q&A
OHLAC 1.5

Balanced all-rounder — work help, tutoring, and everyday writing.

Best for
Work help, tutoring, everyday writing
Speed
Fast
Capability
Balanced
Credits
Low
Image
Image understanding
Tags
Work advicetutoringcreative writing
OHLAC Pro

Better at hard problems — important decisions, structuring plans, and long documents.

Best for
Important decisions, structuring plans, long documents, document reading
Speed
Medium
Capability
Deep
Credits
Medium
Image
Image understanding (doc/PDF text extract)
Files
PDF & docs (OCR text extract)
Tags
Complex analysiswork advicelong documents
GPT-5.5 Instant

Quick replies — daily Q&A, light editing, and everyday chat.

Best for
Instant Q&A, light editing, everyday chat
Speed
Fast
Capability
Balanced+
Credits
High
Image
Image understanding
Tags
Daily Q&Awork advicelife advice
Claude Sonnet 4.6

Steady and versatile — work help, tutoring, and coding.

Best for
Work help, tutoring, coding
Speed
Medium
Capability
Balanced
Credits
Medium
Image
Not available
Tags
Work advicetutoringcoding
Gemini 3.1 Pro Preview

Strong on complex material — tutoring, long documents, and deep analysis.

Best for
Long document analysis, tutoring, complex analysis
Speed
Medium
Capability
Deep
Credits
Medium
Image
Image understanding
Tags
Tutoringlong documentscomplex analysis

How to choose

Match by task type (suggestions, not rules):

Everyday use

GPT-5.6 LunaGemini 3.5 FlashGemini 3.7 FlashGrok 4.5OHLAC 1.0OHLAC 1.5

Hard problems / decisions

Claude Fable 5GPT-5.6 SolClaude Opus 4.8Grok 4.6

Writing

OHLAC 1.5GPT-5.6 TerraClaude Sonnet 5Grok 4.5Grok 4.6

Coding

GPT-5.6 TerraClaude Sonnet 5Claude Opus 4.8

Tutoring / research

Gemini 3.1 Pro PreviewClaude Opus 4.7Gemini 3.5 Flash

Image understanding

Most models support imagesOHLAC Pro (doc/PDF OCR)

Open dialogue

Grok 4.5Grok 4.6

Principle

Start with the most affordable model that fits the task, and upgrade only when needed. Credit rules →

Notes

  • Speed: relative only (very fast > fast > medium > slower > slow) — not a performance guarantee.
  • Capability: lightweight < balanced < balanced+ < deep < flagship.
  • Credit level: relative guidance; settlement follows real server usage.
  • Image / file: only capabilities marked available are open to users.
  • OHLAC Auto: planned, not live yet — do not treat as automatic routing.

Help & Docs

モデル説明

フラッグシップ / 準フラッグシップ / コスパ重視の分類で、位置づけと選び方を確認。

バージョン
0.3.0
施行日
2026年8月10日
最終更新
2026年9月1日

Explore OHLAC 默幕 models by product badge — so you can quickly see what each model is for and how much it tends to cost.

Credit cost is shown as very low / low / medium / high / very high for relative guidance only. Final settlement follows actual server-side usage. Credit rules →

Model overview

Flagship

Claude Fable 5 Flagship

Flagship creation and analysis — complex expression, long documents, and high-bar writing.

Best for
Top-tier creation, deep analysis, complex writing
Speed
Slow
Capability
Flagship
Credits
Very high
Image
Not available
Tags
Creative writingcomplex analysislong documents
GPT-5.6 Sol Flagship

Strongest reasoning model — important decisions, planning, and complex analysis.

Best for
Important decisions, complex analysis, work planning
Speed
Slower
Capability
Deep
Credits
High
Image
Image understanding
Tags
Complex analysiswork advice
Claude Opus 4.8 Flagship

Excels at hard problems — planning, complex analysis, and long documents.

Best for
Complex analysis, long documents, structuring plans
Speed
Slower
Capability
Deep
Credits
High
Image
Image understanding
Tags
Complex analysiswork advicelong documents
Grok 4.6 Flagship

Flagship open thinking — important decisions, complex analysis, and creative writing.

Best for
Important decisions, open discussion, complex analysis
Speed
Medium
Capability
Balanced+
Credits
Low
Image
Image understanding
Tags
Complex analysiscreative writingwork advice

Near-flagship

GPT-5.6 Terra Near-flagship

Strong all-rounder — daily chat, work help, and coding.

Best for
Daily chat, writing, coding
Speed
Medium
Capability
Balanced
Credits
Medium
Image
Image understanding
Tags
Work advicecodingdaily Q&A
Claude Opus 4.7 Near-flagship

Makes complex content clear — tutoring, long documents, and professional writing.

Best for
Tutoring, long documents, professional writing
Speed
Slower
Capability
Deep
Credits
High
Image
Image understanding
Tags
Learningcreative writinglong documents
Claude Opus 4.6 Near-flagship

Steady on hard problems — comparing options, organizing material, and expert Q&A.

Best for
Comparing options, organizing material, expert Q&A
Speed
Slower
Capability
Deep
Credits
High
Image
Image understanding
Tags
Work advicecomplex analysistutoring
Claude Sonnet 5 Near-flagship

Strong all-rounder — work help, daily chat, and coding.

Best for
Work help, daily chat, coding
Speed
Medium
Capability
Balanced+
Credits
Medium
Image
Image understanding
Tags
Work advicecodingdaily Q&A

Great value

GPT-5.6 Luna Great value

Light and easy — daily Q&A, organizing info, and life questions.

Best for
Quick Q&A, organizing info, life questions
Speed
Fast
Capability
Lightweight
Credits
Low
Image
Image understanding
Tags
Daily Q&Alife advice
Gemini 3.5 Flash Great value

Fast and efficient — daily Q&A, tutoring, and organizing material.

Best for
Daily Q&A, tutoring, organizing material
Speed
Fast
Capability
Lightweight
Credits
Low
Image
Image understanding
Tags
Daily Q&Atutoringlong documents
Gemini 3.7 Flash Great value

Fast and flexible — daily Q&A, brainstorming, and everyday organizing.

Best for
Daily Q&A, brainstorming, everyday organizing
Speed
Fast
Capability
Lightweight
Credits
Low
Image
Image understanding
Tags
Daily Q&Acreative writingwork advice
Grok 4.5 Great value

Balanced with personality — open dialogue, creative writing, and work discussion.

Best for
Open dialogue, creative writing, work discussion
Speed
Medium
Capability
Balanced
Credits
Low
Image
Image understanding
Tags
Creative writingwork adviceconversation

Other models

OHLAC 1.0

Light and reliable — daily chat, life questions, and simple rewriting.

Best for
Daily chat, life questions, simple rewriting
Speed
Very fast
Capability
Lightweight
Credits
Low
Image
Image understanding
Tags
Life advicedaily Q&A
OHLAC 1.5

Balanced all-rounder — work help, tutoring, and everyday writing.

Best for
Work help, tutoring, everyday writing
Speed
Fast
Capability
Balanced
Credits
Low
Image
Image understanding
Tags
Work advicetutoringcreative writing
OHLAC Pro

Better at hard problems — important decisions, structuring plans, and long documents.

Best for
Important decisions, structuring plans, long documents, document reading
Speed
Medium
Capability
Deep
Credits
Medium
Image
Image understanding (doc/PDF text extract)
Files
PDF & docs (OCR text extract)
Tags
Complex analysiswork advicelong documents
GPT-5.5 Instant

Quick replies — daily Q&A, light editing, and everyday chat.

Best for
Instant Q&A, light editing, everyday chat
Speed
Fast
Capability
Balanced+
Credits
High
Image
Image understanding
Tags
Daily Q&Awork advicelife advice
Claude Sonnet 4.6

Steady and versatile — work help, tutoring, and coding.

Best for
Work help, tutoring, coding
Speed
Medium
Capability
Balanced
Credits
Medium
Image
Not available
Tags
Work advicetutoringcoding
Gemini 3.1 Pro Preview

Strong on complex material — tutoring, long documents, and deep analysis.

Best for
Long document analysis, tutoring, complex analysis
Speed
Medium
Capability
Deep
Credits
Medium
Image
Image understanding
Tags
Tutoringlong documentscomplex analysis

How to choose

Match by task type (suggestions, not rules):

Everyday use

GPT-5.6 LunaGemini 3.5 FlashGemini 3.7 FlashGrok 4.5OHLAC 1.0OHLAC 1.5

Hard problems / decisions

Claude Fable 5GPT-5.6 SolClaude Opus 4.8Grok 4.6

Writing

OHLAC 1.5GPT-5.6 TerraClaude Sonnet 5Grok 4.5Grok 4.6

Coding

GPT-5.6 TerraClaude Sonnet 5Claude Opus 4.8

Tutoring / research

Gemini 3.1 Pro PreviewClaude Opus 4.7Gemini 3.5 Flash

Image understanding

Most models support imagesOHLAC Pro (doc/PDF OCR)

Open dialogue

Grok 4.5Grok 4.6

Principle

Start with the most affordable model that fits the task, and upgrade only when needed. Credit rules →

Notes

  • Speed: relative only (very fast > fast > medium > slower > slow) — not a performance guarantee.
  • Capability: lightweight < balanced < balanced+ < deep < flagship.
  • Credit level: relative guidance; settlement follows real server usage.
  • Image / file: only capabilities marked available are open to users.
  • OHLAC Auto: planned, not live yet — do not treat as automatic routing.

Help & Docs

Modelos

Agrupados por flagship / near-flagship / gran valor: posicionamiento, capacidades y cómo elegir.

Versión
0.3.0
Fecha de vigencia
10/8/2026
Última actualización
1/9/2026

Explore OHLAC 默幕 models by product badge — so you can quickly see what each model is for and how much it tends to cost.

Credit cost is shown as very low / low / medium / high / very high for relative guidance only. Final settlement follows actual server-side usage. Credit rules →

Model overview

Flagship

Claude Fable 5 Flagship

Flagship creation and analysis — complex expression, long documents, and high-bar writing.

Best for
Top-tier creation, deep analysis, complex writing
Speed
Slow
Capability
Flagship
Credits
Very high
Image
Not available
Tags
Creative writingcomplex analysislong documents
GPT-5.6 Sol Flagship

Strongest reasoning model — important decisions, planning, and complex analysis.

Best for
Important decisions, complex analysis, work planning
Speed
Slower
Capability
Deep
Credits
High
Image
Image understanding
Tags
Complex analysiswork advice
Claude Opus 4.8 Flagship

Excels at hard problems — planning, complex analysis, and long documents.

Best for
Complex analysis, long documents, structuring plans
Speed
Slower
Capability
Deep
Credits
High
Image
Image understanding
Tags
Complex analysiswork advicelong documents
Grok 4.6 Flagship

Flagship open thinking — important decisions, complex analysis, and creative writing.

Best for
Important decisions, open discussion, complex analysis
Speed
Medium
Capability
Balanced+
Credits
Low
Image
Image understanding
Tags
Complex analysiscreative writingwork advice

Near-flagship

GPT-5.6 Terra Near-flagship

Strong all-rounder — daily chat, work help, and coding.

Best for
Daily chat, writing, coding
Speed
Medium
Capability
Balanced
Credits
Medium
Image
Image understanding
Tags
Work advicecodingdaily Q&A
Claude Opus 4.7 Near-flagship

Makes complex content clear — tutoring, long documents, and professional writing.

Best for
Tutoring, long documents, professional writing
Speed
Slower
Capability
Deep
Credits
High
Image
Image understanding
Tags
Learningcreative writinglong documents
Claude Opus 4.6 Near-flagship

Steady on hard problems — comparing options, organizing material, and expert Q&A.

Best for
Comparing options, organizing material, expert Q&A
Speed
Slower
Capability
Deep
Credits
High
Image
Image understanding
Tags
Work advicecomplex analysistutoring
Claude Sonnet 5 Near-flagship

Strong all-rounder — work help, daily chat, and coding.

Best for
Work help, daily chat, coding
Speed
Medium
Capability
Balanced+
Credits
Medium
Image
Image understanding
Tags
Work advicecodingdaily Q&A

Great value

GPT-5.6 Luna Great value

Light and easy — daily Q&A, organizing info, and life questions.

Best for
Quick Q&A, organizing info, life questions
Speed
Fast
Capability
Lightweight
Credits
Low
Image
Image understanding
Tags
Daily Q&Alife advice
Gemini 3.5 Flash Great value

Fast and efficient — daily Q&A, tutoring, and organizing material.

Best for
Daily Q&A, tutoring, organizing material
Speed
Fast
Capability
Lightweight
Credits
Low
Image
Image understanding
Tags
Daily Q&Atutoringlong documents
Gemini 3.7 Flash Great value

Fast and flexible — daily Q&A, brainstorming, and everyday organizing.

Best for
Daily Q&A, brainstorming, everyday organizing
Speed
Fast
Capability
Lightweight
Credits
Low
Image
Image understanding
Tags
Daily Q&Acreative writingwork advice
Grok 4.5 Great value

Balanced with personality — open dialogue, creative writing, and work discussion.

Best for
Open dialogue, creative writing, work discussion
Speed
Medium
Capability
Balanced
Credits
Low
Image
Image understanding
Tags
Creative writingwork adviceconversation

Other models

OHLAC 1.0

Light and reliable — daily chat, life questions, and simple rewriting.

Best for
Daily chat, life questions, simple rewriting
Speed
Very fast
Capability
Lightweight
Credits
Low
Image
Image understanding
Tags
Life advicedaily Q&A
OHLAC 1.5

Balanced all-rounder — work help, tutoring, and everyday writing.

Best for
Work help, tutoring, everyday writing
Speed
Fast
Capability
Balanced
Credits
Low
Image
Image understanding
Tags
Work advicetutoringcreative writing
OHLAC Pro

Better at hard problems — important decisions, structuring plans, and long documents.

Best for
Important decisions, structuring plans, long documents, document reading
Speed
Medium
Capability
Deep
Credits
Medium
Image
Image understanding (doc/PDF text extract)
Files
PDF & docs (OCR text extract)
Tags
Complex analysiswork advicelong documents
GPT-5.5 Instant

Quick replies — daily Q&A, light editing, and everyday chat.

Best for
Instant Q&A, light editing, everyday chat
Speed
Fast
Capability
Balanced+
Credits
High
Image
Image understanding
Tags
Daily Q&Awork advicelife advice
Claude Sonnet 4.6

Steady and versatile — work help, tutoring, and coding.

Best for
Work help, tutoring, coding
Speed
Medium
Capability
Balanced
Credits
Medium
Image
Not available
Tags
Work advicetutoringcoding
Gemini 3.1 Pro Preview

Strong on complex material — tutoring, long documents, and deep analysis.

Best for
Long document analysis, tutoring, complex analysis
Speed
Medium
Capability
Deep
Credits
Medium
Image
Image understanding
Tags
Tutoringlong documentscomplex analysis

How to choose

Match by task type (suggestions, not rules):

Everyday use

GPT-5.6 LunaGemini 3.5 FlashGemini 3.7 FlashGrok 4.5OHLAC 1.0OHLAC 1.5

Hard problems / decisions

Claude Fable 5GPT-5.6 SolClaude Opus 4.8Grok 4.6

Writing

OHLAC 1.5GPT-5.6 TerraClaude Sonnet 5Grok 4.5Grok 4.6

Coding

GPT-5.6 TerraClaude Sonnet 5Claude Opus 4.8

Tutoring / research

Gemini 3.1 Pro PreviewClaude Opus 4.7Gemini 3.5 Flash

Image understanding

Most models support imagesOHLAC Pro (doc/PDF OCR)

Open dialogue

Grok 4.5Grok 4.6

Principle

Start with the most affordable model that fits the task, and upgrade only when needed. Credit rules →

Notes

  • Speed: relative only (very fast > fast > medium > slower > slow) — not a performance guarantee.
  • Capability: lightweight < balanced < balanced+ < deep < flagship.
  • Credit level: relative guidance; settlement follows real server usage.
  • Image / file: only capabilities marked available are open to users.
  • OHLAC Auto: planned, not live yet — do not treat as automatic routing.
© 2026 OHLAC
说明中心 模型说明积分与扣费