Coding Stacks
Mix and match AI coding plans to maximize model coverage within your budget. Each bundle covers IDE, elite models, and volume.
Codex (OpenAI)
Plus
GPT-5.5 standard + thinking model
Expanded Codex access, projects, tasks. 3,000 messages/week for thinking model
OpenCode Go
Go
Kimi K2.5, Kimi K2.6, GLM-5, GLM-5.1, MiMo-V2-Pro, MiMo-V2-Omni, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.5 Plus, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro, DeepSeek V4 Flash
Per 5h: GLM-5.1 880 | GLM-5 1,150 | Kimi K2.5 1,850 | Kimi K2.6 1,150 | MiMo-V2-Pro 1,290 | MiMo-V2-Omni 2,150 | MiMo-V2.5-Pro 1,290 | MiMo-V2.5 2,150 | MiniMax M2.7 3,400 | MiniMax M2.5 6,300 | Qwen3.6 Plus 3,300 | Qwen3.5 Plus 10,200 | DeepSeek V4 Pro 1,300 | DeepSeek V4 Flash 7,450
Codex Plus gets you GPT-5.5 (latest agentic model) for $20. OpenCode Go adds 14 more coding models for $10 including DeepSeek V4 Pro and Flash. This is API + CLI coding only — no IDE integration at this price.
GitHub Copilot
Pro+
Claude Opus 4.7, Claude Sonnet 4/4.5/4.6, GPT-5.2/5.3/5.4/5.4 mini, Gemini 2.5/3/3.1, Raptor mini
1,500 premium requests/mo, unlimited agent mode, unlimited completions
OpenCode Go
Go
Kimi K2.5, Kimi K2.6, GLM-5, GLM-5.1, MiMo-V2-Pro, MiMo-V2-Omni, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.5 Plus, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro, DeepSeek V4 Flash
Per 5h: GLM-5.1 880 | GLM-5 1,150 | Kimi K2.5 1,850 | Kimi K2.6 1,150 | MiMo-V2-Pro 1,290 | MiMo-V2-Omni 2,150 | MiMo-V2.5-Pro 1,290 | MiMo-V2.5 2,150 | MiniMax M2.7 3,400 | MiniMax M2.5 6,300 | Qwen3.6 Plus 3,300 | Qwen3.5 Plus 10,200 | DeepSeek V4 Pro 1,300 | DeepSeek V4 Flash 7,450
StepFun
Flash Mini
Step 3.5 Flash (196B total, 11B active)
~1,500 requests per 5 hours (~22,500 req/day)
Copilot Pro+ is your IDE backbone with Claude Opus 4.7 and 1,500 premium requests. OpenCode adds 14 models for API experiments including DeepSeek V4. StepFun Flash Mini is cheap overflow when Copilot limits hit.
GitHub Copilot
Pro+
Claude Opus 4.7, Claude Sonnet 4/4.5/4.6, GPT-5.2/5.3/5.4/5.4 mini, Gemini 2.5/3/3.1, Raptor mini
1,500 premium requests/mo, unlimited agent mode, unlimited completions
OpenCode Go
Go
Kimi K2.5, Kimi K2.6, GLM-5, GLM-5.1, MiMo-V2-Pro, MiMo-V2-Omni, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.5 Plus, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro, DeepSeek V4 Flash
Per 5h: GLM-5.1 880 | GLM-5 1,150 | Kimi K2.5 1,850 | Kimi K2.6 1,150 | MiMo-V2-Pro 1,290 | MiMo-V2-Omni 2,150 | MiMo-V2.5-Pro 1,290 | MiMo-V2.5 2,150 | MiniMax M2.7 3,400 | MiniMax M2.5 6,300 | Qwen3.6 Plus 3,300 | Qwen3.5 Plus 10,200 | DeepSeek V4 Pro 1,300 | DeepSeek V4 Flash 7,450
Alibaba Cloud
Pro
Qwen3.5-Plus, Qwen3.6-Plus, Qwen3-Coder-Next/Plus, Kimi K2.5, GLM-5, GLM-4.7, MiniMax-M2.5
90,000 requests per month
Alibaba Cloud Pro is the secret weapon — $49 for 90K requests across 7 models including Kimi K2.5, GLM-5, and MiniMax. This gives you the widest model variety under $100.
GitHub Copilot
Pro+
Claude Opus 4.7, Claude Sonnet 4/4.5/4.6, GPT-5.2/5.3/5.4/5.4 mini, Gemini 2.5/3/3.1, Raptor mini
1,500 premium requests/mo, unlimited agent mode, unlimited completions
Codex (OpenAI)
Plus
GPT-5.5 standard + thinking model
Expanded Codex access, projects, tasks. 3,000 messages/week for thinking model
Alibaba Cloud
Pro
Qwen3.5-Plus, Qwen3.6-Plus, Qwen3-Coder-Next/Plus, Kimi K2.5, GLM-5, GLM-4.7, MiniMax-M2.5
90,000 requests per month
OpenCode Go
Go
Kimi K2.5, Kimi K2.6, GLM-5, GLM-5.1, MiMo-V2-Pro, MiMo-V2-Omni, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.5 Plus, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro, DeepSeek V4 Flash
Per 5h: GLM-5.1 880 | GLM-5 1,150 | Kimi K2.5 1,850 | Kimi K2.6 1,150 | MiMo-V2-Pro 1,290 | MiMo-V2-Omni 2,150 | MiMo-V2.5-Pro 1,290 | MiMo-V2.5 2,150 | MiniMax M2.7 3,400 | MiniMax M2.5 6,300 | Qwen3.6 Plus 3,300 | Qwen3.5 Plus 10,200 | DeepSeek V4 Pro 1,300 | DeepSeek V4 Flash 7,450
StepFun
Flash Plus
Step 3.5 Flash (196B total, 11B active)
~6,000 requests per 5 hours (~90,000 req/day)
Everything from the $100 bundle plus dedicated GPT-5.5 via Codex Plus and StepFun Flash Plus for volume. Covers IDE (Copilot), frontier reasoning (Codex GPT-5.5), massive variety (Alibaba), and overflow (StepFun).
GitHub Copilot
Pro+
Claude Opus 4.7, Claude Sonnet 4/4.5/4.6, GPT-5.2/5.3/5.4/5.4 mini, Gemini 2.5/3/3.1, Raptor mini
1,500 premium requests/mo, unlimited agent mode, unlimited completions
Codex (OpenAI)
Pro
GPT-5.4 Pro (5× usage)
5× more usage than Plus. GPT-5.4 Pro, unlimited GPT-5.3, max Codex
Alibaba Cloud
Pro
Qwen3.5-Plus, Qwen3.6-Plus, Qwen3-Coder-Next/Plus, Kimi K2.5, GLM-5, GLM-4.7, MiniMax-M2.5
90,000 requests per month
OpenCode Go
Go
Kimi K2.5, Kimi K2.6, GLM-5, GLM-5.1, MiMo-V2-Pro, MiMo-V2-Omni, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.5 Plus, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro, DeepSeek V4 Flash
Per 5h: GLM-5.1 880 | GLM-5 1,150 | Kimi K2.5 1,850 | Kimi K2.6 1,150 | MiMo-V2-Pro 1,290 | MiMo-V2-Omni 2,150 | MiMo-V2.5-Pro 1,290 | MiMo-V2.5 2,150 | MiniMax M2.7 3,400 | MiniMax M2.5 6,300 | Qwen3.6 Plus 3,300 | Qwen3.5 Plus 10,200 | DeepSeek V4 Pro 1,300 | DeepSeek V4 Flash 7,450
Codex Pro ($100) gets you GPT-5.4 Pro at 5× usage — the strongest Codex tier. Paired with Alibaba's 90K req/mo for volume and Copilot Pro+ for IDE. Best performance-per-dollar at $200.
Claude
Max 5x
All Claude models (Opus, Sonnet, Haiku)
5× Pro capacity. Dynamic 5-hour rolling window. Priority peak access.
GitHub Copilot
Pro+
Claude Opus 4.7, Claude Sonnet 4/4.5/4.6, GPT-5.2/5.3/5.4/5.4 mini, Gemini 2.5/3/3.1, Raptor mini
1,500 premium requests/mo, unlimited agent mode, unlimited completions
OpenCode Go
Go
Kimi K2.5, Kimi K2.6, GLM-5, GLM-5.1, MiMo-V2-Pro, MiMo-V2-Omni, MiMo-V2.5-Pro, MiMo-V2.5, Qwen3.5 Plus, Qwen3.6 Plus, MiniMax M2.5, MiniMax M2.7, DeepSeek V4 Pro, DeepSeek V4 Flash
Per 5h: GLM-5.1 880 | GLM-5 1,150 | Kimi K2.5 1,850 | Kimi K2.6 1,150 | MiMo-V2-Pro 1,290 | MiMo-V2-Omni 2,150 | MiMo-V2.5-Pro 1,290 | MiMo-V2.5 2,150 | MiniMax M2.7 3,400 | MiniMax M2.5 6,300 | Qwen3.6 Plus 3,300 | Qwen3.5 Plus 10,200 | DeepSeek V4 Pro 1,300 | DeepSeek V4 Flash 7,450
Claude Max 5x is native Claude at the highest tier — 5× Pro capacity for deep reasoning and long context. Copilot Pro+ covers IDE + other frontier models. OpenCode fills volume and variety with 14 models including DeepSeek V4. If Claude is your primary model, this is the definitive bundle.