返回资讯中心
外部精选
软件工程
#开发者实时流

Free-AI-tools – Curated list of free and low cost AI tools

Curated list of free LLM APIs, coding copilots, AI IDEs, agents, and infrastructure tools for building real AI applications.

Hacker News 最新peter_d_sherman2026/10/15 · 2026/10/1530 分钟阅读

以下正文同步自 Hacker News 最新,版权归原站所有,已转换为易读排版。

Curated list of free LLM APIs, coding copilots, AI IDEs, agents, and infrastructure tools for building real AI applications.

✅ Free GPT-6 / Claude 5.5 / Gemini 3.8API access

🤖 Coding copilots and AI-native IDEs (Cursor, Trae, Windsurf)

💰 Cheapest AI APIs ($0.08-0.50 per 1M tokens)

📚 RAG stack tools (vector DBs, embeddings, frameworks)

🎯 Agent frameworks and automation tools

🔒 Local models for privacy (Ollama, Llama, Qwen)

🏗️ Production-ready stack configurations

🆕 Claude Sonnet 5.5, Claude Opus 5.5, Haiku 5.5 — GPT-6 Astra, GPT-6.1 Sol, GPT-6 Luna — gemini 3.8 Flash, Gemini 4 Argon — GitHub Copilot AI Credits — Windsurf Max — Trae Ultra — OpenCode 167k⭐ — Xiaomi MiMo V2.5 Pro

Goal: Help developers build AI apps without paying $200/month.

Please don't abuse these services, else we might lose them for everyone. The number becomes 550+ when you add all the models and sub services of all the tools provided. When raising issues or pull requests please dont add your own paid, expensive personal projects.

October 2026 Model Generation Leap: Major providers have transitioned to next-generation architectures: OpenAI introduced the GPT-6 family (GPT-6 Astra, GPT-6.1 Sol, GPT-6 Luna); Anthropic deployed the Claude 5.5 generation (Claude Opus 5.5, Claude Sonnet 5.5, Claude Haiku 5.5); Google rolled out gemini 3.8 Flash / Flash-Lite as the default workhorse and previewed Gemini 4 Argon. The new baseline for high-throughput coding and daily development is Claude Sonnet 5.5, GPT-6.1 Sol, and gemini 3.8 Flash, while complex multi-hour reasoning and autonomous sandboxing target GPT-6 Astra and Claude Opus 5.5.

Pricing & Billing Updates: Windsurf switched to a quota-based model (Pro $20, Teams $40, new Max $200) on Mar 18. Trae moved to a 5-tier token system (Lite $3, Pro $10, Pro+ $30, Ultra $100) on Feb 24. Qoder's launch promo ended — standard standalone pricing settled at Pro $30/mo (2,000 credits, or CNY 59/mo on Qoder CN All-in-One), Teams $40/seat (3,000 credits). Cursor split Teams into Teams Standard ($40/seat) & Teams Premium ($120/seat), credited $70 third-party usage on Pro+ ($60/mo), and shifted Bugbot to ~$1.00–$1.50/run. GitHub Copilot summer credit promotion ended Sep 1, 2026 (reverting to standard $19 / 1,900 credits for Business and $39 / 3,900 credits for Enterprise). Xiaomi MiMo V2.5 API permanent pricing schedule flattened to $1.00/1M input, $3.00/1M output, and $0.20/1M cached across all context lengths up to 1M tokens.

Unlike: awesome-ai (general list), ai-collection (marketing focus), toolify (affiliate-heavy)

This is for: Builders who want to ship AI features this week.

If this repo helped you build something or saved you money:

⭐ Star this repo — it helps more builders discover free AI resources.

[🔄 Share with your team] — spread the knowledge.

📝 Contribute — found a new free tier? Updated pricing? PRs welcome!

Quick Comparison

Free LLM API Providers

Fully Free Providers

Providers with Trial Credits

AI-Powered IDEs

IDEs with Pro-Grade Models

IDEs with Basic Models

CLI Coding Tools

CLI Tools with Pro-Grade Models

CLI Tools with Basic Models

API Providers for AI Coding Tools

Paid Tiers Comparison

Local Models

free-coding-models CLI

Additional 2026 AI Tools

Agentic Workflow Platforms

Data Visualization & Analysis

Creative & Multimedia Tools

Productivity & Research Tools

Vertical AI

Marketing & SEO Tools

Open Source & Local Tools

🏗️ Recommended Stacks

⚡ Realtime & Streaming APIs

🎙️ Speech Models

🎨 Image Generation Models

🎬 Video Generation APIs

🌐 AI Browser Automation

💾 Cheap Vector DB Hosting

🏛️ Common AI Architecture Patterns

💵 Model Price Comparison

🎯 Best Models by Use Case

⏱️ Rate Limit Comparison

✅ Commercial Use Summary

🧩 RAG Stack Tools

🔢 Best Free Embedding APIs

🖥️ AI Hosting & GPU Providers

📊 AI Evaluation Tools

📐 Structured Output Tools

🏷️ Legend

Contributing

License

Quick Comparison

Free LLM API Providers Summary

Provider Models Free Tier Credit Card

NVIDIA NIM 46 40 req/min No

OpenRouter 25 50/day (1K/day with $10) No

Groq 20+ 1K-14.4K req/day No

Google AI Studio 9 5-500 req/day No

Cloudflare Workers AI 47+ 10K neurons/day No

Cerebras 4 1.5M tokens/day No

Mistral La Plateforme 10+ 1B tokens/month No

AI-Powered IDEs with Free Pro-Grade Access

IDE Pro-grade Models Free Tier Limit Credit Card

Cursor Claude Sonnet 5.5 / GPT-6.1 Sol / Custom Limited free tier (Hobby) No

Trae Doubao / Claude 3.5/3.7 / GPT-4o (Region-dependent) 5,000 auto-completions/month No

Windsurf Claude Sonnet 5.5 / Opus 5.5, GPT-6.1 Sol Light quota (daily/weekly) No

Qoder Qwen3.6-Plus, Qwen3-Coder-480B, GPT-6.1 Sol Unlimited completions + limited chat No

AI GitHub PR Review Tools

Tool Starting Price Free Tier Features Credit Card

PrixAI Free / $10 paid plan Free trial available Unlimited reviews Auto-fix PRs, issue planning No

Bito Free / $25 paid plans Free trial available AI PR reviews/Unlimited reviews No

Sourcery ~$12/month Free trial available Code quality reviews No

CLI Coding Tools with Free Pro-Grade Access

Tool Pro-grade Models Free Tier Limit Credit Card

Gemini CLI gemini 3.8 Flash / Flash-Lite 1,500 req/day No

Rovo Dev CLI Claude Sonnet 5.5, GPT-6.1 Sol Enterprise Cloud add-on (5M tok/day) Paid Atlassian Cloud org required

Warp GPT-6.1 Sol, Claude Sonnet 5.5 150 credits/mo (first 2 mo), 75/mo after No

GitHub Copilot Claude Sonnet 5.5, GPT-6.1 Sol, GPT-6 Astra 50 chat + 2K completions/month No

Jules gemini 3.8 Flash / Gemini 3 Pro 15 tasks/day No

Amazon Q Developer Claude 3.5 Sonnet (AWS Bedrock) 50 interactions/month (AWS Builder ID) No

OpenCode 75+ providers (BYOK) + Go bundle Free (Zen) / Go $10/mo No

Xiaomi MiMo MiMo-V2.5-Pro, MiMo-V2.5, MiMo-V2-Omni Free API credits No

ForgeCode 300+ models via OpenRouter 10K tokens/day No

RooCode Bring your own keys Unlimited (BYOK) No

Goose Bring your own keys Unlimited (BYOK) No

OhMyPi Bring your own keys Unlimited (BYOK) No

What Qualifies as "Pro-Grade"?

Models achieving ≥60% on SWE-bench Verified / Pro or leading agentic benchmarks (Terminal-Bench 4.0, DeepSWE v1.1):

Note: [verify] indicates scores need verification from official sources. Always check current benchmarks before making decisions.

Ready-made combinations for different use cases. Copy-paste these configurations.

Layer Tool Why

IDE Cursor Hobby / Qoder Limited completions + Claude Sonnet 5.5 / GPT-6.1 Sol chat

CLI Gemini CLI (3.5 Flash) / OpenCode Zen 1,500 req/day Flash, free open-source models

API OpenRouter + Groq 50 req/day + 14.4K req/day combo

Local Ollama + Qwen3.6-Plus Unlimited offline

Automation n8n Self-hosted Unlimited workflows

Vector DB ChromaDB / LanceDB Free local storage

Total Cost: $0/month

Layer Tool Speed

Inference Groq / Cerebras 2,000 tokens/sec (Cerebras)

Coding Qwen3.6-Plus via Groq 1,000 req/day (71.2% SWE)

Agent OpenCode Zen MiniMax M2.5 (80.2%), DeepSeek V4

Cache DeepSeek V4 $0.30/$0.50 per 1M, 90% cache discount

Edge Cloudflare Workers AI Global CDN

Best for: Real-time apps, trading bots, live coding assistants

Layer Tool Cost

IDE Trae Lite $3/mo ($5 basic usage + bonus)

IDE Trae Pro $10/mo ($20 basic usage + bonus, SOLO mode)

API OpenRouter $10 1K req/day + BYOK 1M/month free

CLI OpenCode Free (BYOK) or Go $10/mo

CLI Xiaomi MiMo Lite $6/mo (60M credits, ~120 tasks)

CLI Gemini CLI gemini 3.8 Flash / Flash-Lite

Local Ollama Free

Embeddings Jina AI Free tier

Total Cost: ~$10/month for pro-grade everything

Layer Tool Privacy

Models Ollama + Llama 3.3 / Qwen3-Coder Runs locally

IDE Continue.dev + VS Code BYO local models

CLI Aider + local Ollama Git-integrated, offline

Chat UI Open WebUI Self-hosted ChatGPT alternative

Vector DB ChromaDB / LanceDB Local embeddings storage

Speech Whisper (local) Offline transcription

Best for: Healthcare, legal, finance - any sensitive data

Component Tool Role

Orchestrator n8n / Gumloop Workflow automation

Reasoning DeepSeek R1 / DeepSeek V4 Complex decision making

Execution Qwen3.6-Plus Code generation

Memory ChromaDB / Supabase Vector Long-term context

Embeddings Jina Embeddings v3 (1M tokens/day free) Semantic search

Monitoring LangSmith Trace agent steps

Best for: Autonomous research assistants, code review bots, data processing pipelines

Component Tool Purpose

Framework LlamaIndex / LangChain RAG orchestration

Vector DB ChromaDB / Weaviate / Supabase Document storage

Embeddings E5-Mistral-7B (best accuracy) Text vectorization

Chunking LlamaIndex Smart document splitting

Reranking Cohere Rerank Improve retrieval accuracy

LLM Claude Sonnet 5.5 / GPT-6.1 Sol Answer generation

Eval RAGAS Measure RAG performance

Best for: Document QA, legal document analysis, knowledge bases

Fully Free Providers

OpenRouter

Limits: 20 RPM, 29 free models (262K context max, March 2026), models share quota

Unified API gateway for 100+ LLMs. OpenAI and Anthropic SDK-compatible. China-friendly with Hong Kong direct access (100-300ms latency). No monthly fees, pay per token.

Data is used for training when used outside UK/CH/EEA/EU.

Rate limits: Tier 1 (default): 250 RPD | Tier 2: Requires $250 spend + 30 days

Note: Data training outside UK/CH/EEA/EU still applies.

Phone number verification required. Models tend to be context window limited.

Limits: 1K credits signup, up to 5K total, 40 RPM (phone verify required)

Free tier requires opting into data training; phone verification required

Limits (per-model): 1 req/s, 500K tokens/min, 1B tokens/month

Self-hosted, open-source AI gateway for routing requests across 20+ providers through an OpenAI-compatible API.

AI gateway with curated models. Free models may use data for improvement.

Model Limits

GPT-OSS 120B 30 req/min, 60K tokens/min, 900 req/hour, 1.5M tokens/day

Llama 3.1 8B Same limits as above

Qwen3-235B Available via API

Groq

Model Limits

Llama 3.1 8B 14,400 req/day, 6K tokens/min

Llama 3.3 70B 1,000 req/day, 12K tokens/min

Llama 4 Maverick/Scout 1,000 req/day

Whisper Large v3/v3 Turbo 7,200 audio-sec/min, 2,000 req/day

Qwen3-32B 1,000 req/day, 6K tokens/min

Kimi K2 Instruct 1,000 req/day, 10K tokens/min

GPT-OSS 20B/120B 1,000 req/day, 8K tokens/min

And 15+ more

Cloudflare Workers AI

Limits: 10,000 neurons/day

Provider Credits Duration Notes

Fireworks $1 Permanent Various open models

Baseten $30 Permanent Pay by compute time

Nebius $1 Permanent Various open models

Novita $0.50 1 year Various open models

AI21 $10 3 months Jamba family

Upstage $10 3 months Solar Pro/Mini

NLP Cloud $15 Permanent Phone verification required

Alibaba Cloud 1M tokens/model 90 days Qwen models

Modal $5-30/month Monthly Pay by compute time

Inference.net $1 (+$25 on survey) Permanent Various open models

Hyperbolic $1 Permanent DeepSeek, Llama, Qwen, GPT-OSS

SambaNova Cloud $5 3 months Llama, Qwen, DeepSeek

Scaleway 1M tokens Permanent DeepSeek, Llama, Mistral, Gemma

Vercel AI Gateway $5 credit Monthly Requires active Vercel account

Additional Free API Providers

Provider Models Free Tier Environment Variable

ZAI 7 Free tier (generous quota) ZAI_API_KEY

SiliconFlow 6 1K RPM, 50K TPM SILICONFLOW_API_KEY

OVHcloud AI Endpoints 8 2 req/min (no key), 400 RPM with key OVH_AI_ENDPOINTS_ACCESS_TOKEN

Chutes AI 4 Free community GPU-powered CHUTES_API_KEY

DeepInfra 4 200 concurrent requests DEEPINFRA_API_KEY

Replicate 2 6 req/min (no payment), up to 3K RPM with payment REPLICATE_API_TOKEN

AI-Powered IDEs

Full-featured integrated development environments with built-in AI assistance.

Cursor

Model: Adaptive routing with Claude Sonnet 5.5, Claude Opus 5.5, and GPT-6.1 Sol (accessible via AI Gateway / custom config)

Models: Doubao (natively in China), Claude 3.5/3.7 Sonnet & GPT-4o (international routing)

Models: OpenAI (GPT-6.1 Sol), Anthropic (Claude Sonnet 5.5, Claude Opus 5.5), Google, xAI

Pricing | Pricing Announcement (Mar 18, 2026)

Models: Multi-agent (frontend/backend/testing agents)

Best for:

正文由 探How 从来源站点 RSS 同步,内容未经改写;遇到排版缺失或需要图片、视频时请以原文为准。

打开原文
去社区讨论 →