- 查看我们最新发布的预测引擎：MiroFish-简洁通用的群体智能引擎，预测万物
- <div align="center"> <img src=" alt="banner" width="500"> </div>
- “数据分析三板斧”全线贯通：我们激动的宣布 MiroFish 正式发布！随着最后一块版图补齐，我们构建了从 BettaFish（数据收集与分析）到 MiroFish（全景预测）的完整链路。至此，从原始数据到智能决策的闭环已成，让预见未来成为可能！
- <img width="1502" height="990" alt="04fb364a3122e3ec7dfab8929897745a" src=" />
BettaFish by 666ghj (community project)
Pulse Score
43 / 100
Rank #73 of 258
GitHub stars
42.3K
about +8 a day
Latest release
3.0.0
9mo ago
Price
Free (OSS)
checked Oct 6, 2026
Overview
An open-source multi-agent system that collects and analyses social-media posts and comments across platforms and writes public-opinion research reports; built in Python without an external agent framework.
The GitHub repository has 42.3K stars, 278 of them added in the last 30 days. The latest stable release is 3.0.0, published Dec 23, 2025.
BettaFish usage and attention over time
Download the raw daily series: bettafish.csv
Key facts
| Maker | 666ghj (community project) |
|---|---|
| Type | Open-source multi-agent public-opinion analysis system |
| First released | July 2024 |
| License | GPL-2.0 |
| Main language | Python |
| Open issues and PRs | 4 |
Pricing
Free and open source (GPL-2.0), self-hosted; requires your own model and search API keys.
Source: official pricing page, checked Oct 6, 2026. Prices change often; confirm before you buy.
Pulse Score breakdown
Adoption—
Community68
Attention—
Momentum50
Each component is scored 0 to 100 from public signals; a dash means no data for it. Methodology.
BettaFish news and signals
-
101 agents added to AgentGid
DeepSeek Harness, CopilotKit, Orca, eve, ARTEMIS, Jasper, Harvey, Paradox, Eightfold AI Interviewer, TesterArmy e2e, Copy.ai, Consensus and 89 more
What changed: recent BettaFish releases
- feat(insightagent): Insight Agent新增了DeepSearchAgent.clusterandsampleresults()方法，该方法对搜索结果进行文本编码，使用 KMeans 聚类并按热度排序采样，然后从每个聚类中选取代表性语义返回
- feat(xhs): 集成xhshow库优化签名生成与请求参数
- Feat: 同样的效果更经济的搜索选择，新增安思派AI搜索支持
- Shutdown only clean 增加安全关闭系统的功能
- Update Via main
- @ptghb made their first contribution in
- IR → HTML/PDF 全链路渲染：新增 htmlrenderer 与 pdfrenderer，通过解耦内容与表现层，支持纯命令行模式（reportengineonly.py）生成报告。
- 离线依赖集成：内置思源宋体子集、MathJax、Chart.js 及 html2canvas，无需依赖外部 CDN，确保内网环境下的完美渲染。
- PDF 导出增强：新增 /api/report/export/pdf 接口。导出前自动进行 Pango 依赖环境自检，支持一键生成高质量矢量 PDF。
- Chart.js 安全校验：引入 chartvalidator 结合 LLM 自动修复链路，有效防止非法配置导致的前端崩溃或 XSS 注入风险。
- 数据清洗回归测试：增加 testreportenginesanitization 测试套件，覆盖章节表格的容错修复，保障渲染端健壮性。
- 可视化日志控制台：前端采用双缓冲虚拟列表 + SSE 历史补偿机制，后端提供日志管理接口，大幅提升长任务监控体验。
- 稳定性提升：修复了若干Agent问题，修复了LLM易超时、报错问题。
- 动态主机/端口与外部服务配置：v1.1.0 引入 .env 动态主机地址配置，允许用户自定义主机域名和端口用于网络或镜像访问。在 v1.2.0 中，该机制进一步完善，默认提供基于 pydantic 的配置验证系统，提升部署稳定性。
- LLM 接口管理：新增 LLM 接口独立配置功能，支持多模型、多接口动态切换，解决接口中英文调用冲突问题。通过统一配置可管理 Base URL、Token、模型参数，方便不同模型与服务的无缝接入。
- 界面优化调整：新增前端动态渲染逻辑，支持在 Web UI 中修改系统参数与分析选项。
- Docker 支持增强：新版镜像已全面支持自动化部署。自动创建数据库并支持 .env 参数传入。优化 Dockerfile 构建参数，支持 ARM 架构（Apple M 系列）平台运行。修复 docker-compose.yml 文件路径问题，避免配置遗漏。
- 更新 MediaCrawler 与 ForumEngine 模块，解决通信中断和编码异常。
- 修复：使用动态主机地址支持局域网内后端服务访问
- 改为使用uv安装，更加快速，且修复torch和transformers未安装的问题
- Fixes #61：修复fetchnews无法正常获取的问题
- chore: stop track config.py
- hotfix(database): fix sourcekeyword not in table bilibilivideo
- hotfix(timeout): increase llm call timeout to 1800s
- Insight Agent: 私有数据库挖掘，深度分析AI代理
- Media Agent: 多模态内容分析，具备强大多模态能力的AI代理
- Query Agent: 精准信息搜索，具备国内外网页搜索能力的AI代理
- Report Agent: 智能报告生成，内置模板的多轮报告生成AI代理
- Forum Engine: Agent"论坛"协作机制，实现链式思维碰撞与辩论
- AI爬虫集群7x24小时不间断作业
BettaFish alternatives
| # | Agent | Pulse | Downloads 7d | 7d | 30d | VS Code installs | Stars | Latest release | Price | Last 90 days | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 68 |
ConsensusConsensus |
43 | — | — | — | — | — | — | Free + $20/mo | ||
| 97 |
SciSpace AgentPubGenius Inc. |
37 | — | — | — | — | — | — | Free + $20/mo | ||
| 102 |
ElicitElicit |
36 | — | — | — | — | — | — | Free + $49/mo | — | |
| 141 |
GPT ResearcherAssaf Elovic and contributors |
28 | 13.1K | up 42.4% | down 41.5% | — | 29.9K+28/day | 3.7.010d ago | Free (OSS) | ||
| 152 |
KosmosEdison Scientific |
26 | 13.5K | up 10.2% | up 48.1% | — | — | — | Free + $200/mo | ||
| 199 |
STORMStanford OVAL |
14 | 329 | up 21.0% | down 50.5% | — | 31.6K+9/day | 1.1.020mo ago | Free (OSS) | ||
| No agents match that filter. | |||||||||||
Head to head: BettaFish vs Consensus · BettaFish vs SciSpace Agent · BettaFish vs Elicit · BettaFish vs GPT Researcher
BettaFish FAQ
How much does BettaFish cost?
Free and open source (GPL-2.0), self-hosted; requires your own model and search API keys.
Is BettaFish open source?
Yes. BettaFish is open source, released under the GPL-2.0 license.
How popular is BettaFish?
The GitHub repository has 42.3K stars, 278 of them added in the last 30 days. The latest stable release is 3.0.0, published Dec 23, 2025.
What is the latest version of BettaFish?
The latest stable release we track is 3.0.0, published on Dec 23, 2025.
What are the alternatives to BettaFish?
The closest alternatives in the same category by Pulse Score are Consensus, SciSpace Agent, Elicit, GPT Researcher.
Where these numbers come from
GitHub: 666ghj/bettafish. See data sources for how each one is collected.
Product facts and pricing were checked against: github.com.