Claude Code: Compatible Models & Providers (2025-2026 Edition)

A comprehensive reference guide for developers using Claude Code with official and alternative AI providers. This guide covers the latest flagship models, pricing (Input/Output per 1M tokens), and configuration for Anthropic, Alibaba Qwen, DeepSeek, MiniMax, Moonshot AI (Kimi), Zhipu GLM, and OpenRouter.

💰 Model Pricing Overview (Latest Models)

Provider	Flagship Model	Input ($/1M)	Output ($/1M)	Notes
Anthropic	Claude 4.6 Sonnet	$3.00	$15.00	Best-in-class performance & agentic tasks.
Alibaba	Qwen 3.5	$0.55	$3.50	Next-gen native multimodal, 1M context.
DeepSeek	DeepSeek V3.2	$0.27	$0.42	Extremely efficient, caching enabled.
MiniMax	MiniMax-M2.5	$0.15	$1.20	"Intelligence too cheap to meter", MoE.
Moonshot	Kimi K2.5	$0.60	$3.00	1T parameters MoE, native multimodal.
Zhipu (Z.ai)	GLM-5	$1.00	$3.20	New flagship with strong reasoning.
OpenRouter	Multi-Provider	Varies	Varies	Unified access to all the above.

---

🚀 Provider Setup Guides

Anthropic (Official)

Claude Code officially supports Claude 4.6 Sonnet, which excels at coding tasks and tool use.

Installation:

npm install -g @anthropic-ai/claude-code

Authentication:

• OAuth: Run claude and follow the browser login.
• API Key: Set ANTHROPIC_API_KEY in your environment.

Official Documentation

---

Alibaba Cloud - DashScope (Qwen)

Qwen 3.5 is the latest flagship, supporting massive context windows.

Configuration: Edit ~/.claude/settings.json:

{
  "env": {
    "ANTHROPIC_BASE_URL": "https://dashscope-intl.aliyuncs.com/apps/anthropic",
    "ANTHROPIC_AUTH_TOKEN": "YOUR_DASHSCOPE_API_KEY",
    "ANTHROPIC_MODEL": "qwen3.5-plus",
    "ANTHROPIC_SMALL_FAST_MODEL": "qwen3.5-coder"
  }
}

---

DeepSeek

DeepSeek V3.2 incredibly cost-effective alternatives.

Configuration:

{
  "env": {
    "ANTHROPIC_BASE_URL": "https://api.deepseek.com/anthropic",
    "ANTHROPIC_AUTH_TOKEN": "YOUR_DEEPSEEK_API_KEY",
    "ANTHROPIC_MODEL": "deepseek-reasoner",
    "ANTHROPIC_SMALL_FAST_MODEL": "deepseek-chat"
  }
}

Note: deepseek-reasoner maps to DeepSeek V3.2 (with thinking processes).

---

MiniMax

The new MiniMax-M2.5 series offers the lowest output pricing in its class.

Configuration:

{
  "env": {
    "ANTHROPIC_BASE_URL": "https://api.minimax.io/anthropic",
    "ANTHROPIC_AUTH_TOKEN": "YOUR_MINIMAX_API_KEY",
    "ANTHROPIC_MODEL": "minimax-m2.5",
    "ANTHROPIC_SMALL_FAST_MODEL": "minimax-m2.5-lightning"
  }
}

---

Moonshot AI - Kimi

Kimi K2.5 is a massive 1T-parameter model optimized for long context (256K).

Configuration:

{
  "env": {
    "ANTHROPIC_BASE_URL": "https://api.moonshot.ai/anthropic",
    "ANTHROPIC_AUTH_TOKEN": "YOUR_MOONSHOT_API_KEY",
    "ANTHROPIC_MODEL": "kimi-k2.5",
    "ANTHROPIC_SMALL_FAST_MODEL": "kimi-k2"
  }
}

---

Zhipu (Z.ai) - GLM

GLM-5 is Zhipu's latest flagship model

Configuration:

{
  "env": {
    "ANTHROPIC_BASE_URL": "https://api.z.ai/api/anthropic",
    "ANTHROPIC_AUTH_TOKEN": "YOUR_ZAI_API_KEY",
    "ANTHROPIC_MODEL": "glm-5",
    "ANTHROPIC_SMALL_FAST_MODEL": "glm-4-flash"
  }
}

---

OpenRouter (Multi-Provider)

Best for comparing models or failover. Supports all the above via one token.

Basic Setup:

export ANTHROPIC_BASE_URL="https://openrouter.ai/api"
export ANTHROPIC_AUTH_TOKEN="YOUR_OPENROUTER_KEY"

---

⭐ Special Coding Plans

MiniMax M2.5 Coding Plan

• Max Plan: $50/mo for 1000 prompts/5h window.
• Lightning Edge: Extremely fast generation for repo-wide scans.

GLM Subscription Plans

• Pro: $15/mo, includes Vision and Web Search capabilities.
• Lite: $3/mo, perfect for hobbyists.

---

💡 Quick Tips

1. Thinking Mode: For complex refactors, use models like deepseek-reasoner or kimi-k2-thinking.
2. Context Caching: DeepSeek and Anthropic support caching, which can reduce input costs by up to 90%.
3. Regional Endpoints: If you are in China, use open.bigmodel.cn for GLM or dashscope.aliyuncs.com for Qwen for lower latency.