沧州泊头纯设计网站制作外贸常用网站

前段时间发现了：AI自动化制作web和手机应用软件的利器：bolt.new-CSDN博客

但是它只能使用anthropic的claude-3-5-sonnet-20240620模型，这使它的AI能力大打折扣，那么有办法使用其它模型吗？可以的，bolt.diy可以使用各种AI大模型！官网：https://github.com/stackblitz-labs/bolt.diy

下载软件源代码

git clone https://github.com/stackblitz-labs/bolt.diy

安装和启动

进入bolt.diy目录，然后执行

npm install -g pnpm
pnpm install
pnpm run dev

启动后输出

pnpm run dev> bolt@0.0.7 dev E:\github\bolt.diy
> node pre-start.cjs  && remix vite:dev★═══════════════════════════════════════★B O L T . D I Y⚡️  Welcome  ⚡️
★═══════════════════════════════════════★📍 Current Version Tag: v"0.0.7"
📍 Current Commit Version: "0202aef"Please wait until the URL appears here
★═══════════════════════════════════════★warn  Data fetching is changing to a single fetch in React Router v7
┃ You can use the `v3_singleFetch` future flag to opt-in early.
┃ -> https://remix.run/docs/en/2.13.1/start/future-flags#v3_singleFetch
┗➜  Local:   http://localhost:5173/➜  Network: use --host to expose➜  press h + enter to show help

 登录 http://localhost:5173/

可以看到跟踪信息里有支持的AI模型列表：

 INFO   LLMManager  Registering Provider:  AnthropicINFO   LLMManager  Registering Provider:  CohereINFO   LLMManager  Registering Provider:  DeepseekINFO   LLMManager  Registering Provider:  GoogleINFO   LLMManager  Registering Provider:  GroqINFO   LLMManager  Registering Provider:  HuggingFaceINFO   LLMManager  Registering Provider:  HyperbolicINFO   LLMManager  Registering Provider:  MistralINFO   LLMManager  Registering Provider:  OllamaINFO   LLMManager  Registering Provider:  OpenAIINFO   LLMManager  Registering Provider:  OpenRouterINFO   LLMManager  Registering Provider:  OpenAILikeINFO   LLMManager  Registering Provider:  PerplexityINFO   LLMManager  Registering Provider:  xAIINFO   LLMManager  Registering Provider:  TogetherINFO   LLMManager  Registering Provider:  LMStudioINFO   LLMManager  Registering Provider:  AmazonBedrockINFO   LLMManager  Registering Provider:  Github

配置模型

登录页面后，显示配置页面：

但是我咋没有找到Ollama 、LMStudio或者OpenAI兼容模型配置呢？ 

算了，用OpenAI试试吧。

 明白了，原来是要配置.env文件，官方给了.env.example文件：

# Rename this file to .env once you have filled in the below environment variables!# Get your GROQ API Key here -
# https://console.groq.com/keys
# You only need this environment variable set if you want to use Groq models
GROQ_API_KEY=# Get your HuggingFace API Key here -
# https://huggingface.co/settings/tokens
# You only need this environment variable set if you want to use HuggingFace models
HuggingFace_API_KEY=# Get your Open AI API Key by following these instructions -
# https://help.openai.com/en/articles/4936850-where-do-i-find-my-openai-api-key
# You only need this environment variable set if you want to use GPT models
OPENAI_API_KEY=# Get your Anthropic API Key in your account settings -
# https://console.anthropic.com/settings/keys
# You only need this environment variable set if you want to use Claude models
ANTHROPIC_API_KEY=# Get your OpenRouter API Key in your account settings -
# https://openrouter.ai/settings/keys
# You only need this environment variable set if you want to use OpenRouter models
OPEN_ROUTER_API_KEY=# Get your Google Generative AI API Key by following these instructions -
# https://console.cloud.google.com/apis/credentials
# You only need this environment variable set if you want to use Google Generative AI models
GOOGLE_GENERATIVE_AI_API_KEY=# You only need this environment variable set if you want to use oLLAMA models
# DONT USE http://localhost:11434 due to IPV6 issues
# USE EXAMPLE http://127.0.0.1:11434
OLLAMA_API_BASE_URL=# You only need this environment variable set if you want to use OpenAI Like models
OPENAI_LIKE_API_BASE_URL=# You only need this environment variable set if you want to use Together AI models
TOGETHER_API_BASE_URL=# You only need this environment variable set if you want to use DeepSeek models through their API
DEEPSEEK_API_KEY=# Get your OpenAI Like API Key
OPENAI_LIKE_API_KEY=# Get your Together API Key
TOGETHER_API_KEY=# You only need this environment variable set if you want to use Hyperbolic models
#Get your Hyperbolics API Key at https://app.hyperbolic.xyz/settings
#baseURL="https://api.hyperbolic.xyz/v1/chat/completions"
HYPERBOLIC_API_KEY=
HYPERBOLIC_API_BASE_URL= # Get your Mistral API Key by following these instructions -
# https://console.mistral.ai/api-keys/
# You only need this environment variable set if you want to use Mistral models
MISTRAL_API_KEY=# Get the Cohere Api key by following these instructions -
# https://dashboard.cohere.com/api-keys
# You only need this environment variable set if you want to use Cohere models
COHERE_API_KEY=# Get LMStudio Base URL from LM Studio Developer Console
# Make sure to enable CORS
# DONT USE http://localhost:1234 due to IPV6 issues
# Example: http://127.0.0.1:1234
LMSTUDIO_API_BASE_URL=# Get your xAI API key
# https://x.ai/api
# You only need this environment variable set if you want to use xAI models
XAI_API_KEY=# Get your Perplexity API Key here - 
# https://www.perplexity.ai/settings/api
# You only need this environment variable set if you want to use Perplexity models
PERPLEXITY_API_KEY=# Get your AWS configuration
# https://console.aws.amazon.com/iam/home
# The JSON should include the following keys:
#   - region: The AWS region where Bedrock is available.
#   - accessKeyId: Your AWS access key ID.
#   - secretAccessKey: Your AWS secret access key.
#   - sessionToken (optional): Temporary session token if using an IAM role or temporary credentials.
# Example JSON:
# {"region": "us-east-1", "accessKeyId": "yourAccessKeyId", "secretAccessKey": "yourSecretAccessKey", "sessionToken": "yourSessionToken"}
AWS_BEDROCK_CONFIG=# Include this environment variable if you want more logging for debugging locally
VITE_LOG_LEVEL=debug# Get your GitHub Personal Access Token here -
# https://github.com/settings/tokens
# This token is used for:
# 1. Importing/cloning GitHub repositories without rate limiting
# 2. Accessing private repositories
# 3. Automatic GitHub authentication (no need to manually connect in the UI)
# 
# For classic tokens, ensure it has these scopes: repo, read:org, read:user
# For fine-grained tokens, ensure it has Repository and Organization access
VITE_GITHUB_ACCESS_TOKEN=# Specify the type of GitHub token you're using
# Can be 'classic' or 'fine-grained'
# Classic tokens are recommended for broader access
VITE_GITHUB_TOKEN_TYPE=classic# Example Context Values for qwen2.5-coder:32b
# 
# DEFAULT_NUM_CTX=32768 # Consumes 36GB of VRAM
# DEFAULT_NUM_CTX=24576 # Consumes 32GB of VRAM
# DEFAULT_NUM_CTX=12288 # Consumes 26GB of VRAM
# DEFAULT_NUM_CTX=6144 # Consumes 24GB of VRAM
DEFAULT_NUM_CTX=

还是用自建AI服务器，创建.env文件，文件内容 

OPENAI_API_KEY="your_api_key_here"
OPENAI_BASE_URL="http://192.168.1.5:1337/v1"

最终试验下来这样写

OPENAI_LIKE_API_BASE_URL="http://192.168.1.5:1337/v1"
LMSTUDIO_API_BASE_URL="http://192.168.1.5:1337/v1"

 但是配置了还是没起作用啊，而且根本看不到Ollama 、LMStudio或者OpenAI兼容模型啊？

明白了，原来配置文件是.env.production 。修改它，加上前面的两句，还是不行。

说：Click the settings icon in the sidebar to open the settings menu

但是我没有找到这个按钮啊，明白了，原来是因为我用了360浏览器，结果它左边弄了很多广告图标，导致我把最下面的配置图标也当成了广告。

就这样：

最下面这个配置图标，就是bole.diy的，这是大人能干出来的事？ 

点进去，打开本地大模型选项，终于可以配置了Ollama 、LMStudio和OpenAI兼容模型了！

 INFO   api.llmcall  Generating response Provider: OpenAILike, Model: deepseek-v3

这样模型终于配置完成了！

测试

用这个prompt

帮我建一个类似“雷电”打飞机的游戏页面，使用html5，飞机可以左右移动，用左右方向键移动。景物和障碍从上向下滚动。按空格键发射子弹，子弹击中障碍物后，障碍物消失，并得10分。点击“开始游戏”开始，得分达到100分通关，输出“恭喜通关！”

 运行后，卡在这里：

Analysis Complete

Determining Files to Read

不知道咋回事。

但至少，确实自动生成了很多文件