Updated June 2026

AI Coding Plans & Pricing

Side-by-side comparison of direct AI coding providers, aggregators, and IDE tool bundles — with tiered pricing, model access, and unique differentiators.

46+Providers
166Tier Cards
FreeTiers Available

Direct Providers

10
BRAND LITE / ENTRY PRO / STANDARD PLUS / MAX SCALE / ULTRA MODELS NOTES
GeminiGoogle
Free$0/moGemini app, Flash/Pro access, image gen, Deep Research, Canvas, Gems. 50 daily AI credits
AI Plus$7.99/moEnhanced 3.1 Pro, Deep Research, 200 AI credits/mo, 200 GB storage, Gemini in Gmail/Chrome
AI Pro$19.99/moHigher 3.1 Pro access, 1,000 AI credits/mo, 5 TB storage, Jules agent, Antigravity IDE
AI UltraFrom $99.99/mo5x–20x Pro usage. Deep Think, Gemini Spark, 10K–25K Flow credits, Project Genie
Gemini 2.5 FlashGemini 2.5 ProGemini 3.1 ProGemini 3.1 FlashJules
Deep Research, Canvas, Gems, Antigravity
Mistral AIGDPR-compliant
Free$0/moVibe — SOTA models, 500 memories, image gen, projects, 40+ connectors
Pro$14.99/moStudents $5.99/mo. Higher limits, Mistral Vibe coding, 15 GB storage, 1,000 projects
Team$24.99/moPer user. 30 GB storage/user, domain verify, data export, collaboration
No higher tier
Mistral LargeMistral MediumMistral SmallMixtral 8x22BCodestral
European provider, GDPR-native
Xiaomi MiMo20% off-peak
Lite$6/mo4.1B credits. All 8 models incl. MiMo-V2.5-Pro. Works with OpenClaw, Claude Code, OpenCode, KiloCode. 20% off-peak
Standard$16/mo11B credits. 2.7× Lite. All 8 models. Free TTS (limited time)
Pro$50/mo38B credits. 9.3× Lite. 1M context. 20% off-peak 16:00–24:00 UTC
Max$100/mo82B credits. 20× Lite. All models included
MiMo-V2.5-ProMiMo-V2.5MiMo-V2-FlashMiMo-V2-Omni
1M context on Pro/Max. 20% off-peak (16:00–24:00 UTC)
StepFunMulti-device
Flash Mini$6.99/mo~1,500 req/5h. Multi-device login. Concurrent agents
Flash Plus$9.99/mo~6,000 req/5h. Priority API. 4× Flash Mini
Flash Pro$29/mo~22,500 req/5h. Priority API. 15× Flash Mini. 300+ tok/sec
Flash Max$99/mo~75,000 req/5h. Priority API. 50× Flash Mini
Step 3.5 FlashStep 3.5 PlusStep 3.5 ProStep 3.7 FlashStep 3.7 PlusStep 3.7 Pro
Rolling 5-hour windows. DeepSeek-compatible API
CodexOpenAI
Go$8/moMore GPT-5.5 Instant access, longer memory. May include ads
Plus$20/moGPT-5.5 Thinking, expanded context, deep research, projects, tasks
ProFrom $100/mo5×–20× more usage. GPT-5.5 Pro, unlimited GPT-5.3, max Codex tasks
No separate Ultra tier
GPT-5.5GPT-5.4GPT-5.3o3o3-mini
Official OpenAI coding agent CLI
xAI GrokxAI
SuperGrok Lite$10/moGrok 4, Imagine image gen, limited web + X search
SuperGrok$30/moFull Grok 4 & 4.1, DeepSearch, Imagine, 128K context, 1K text prompts/mo
SuperGrok Heavy$300/moGrok 4 Heavy preview, 256K–428K context, max compute priority, unlimited text
No higher tier
Grok 4Grok 4.1Grok 4 HeavyGrok Code Fast 1
Consumer coding + search suite
Z.AI GLMZhipu AI
Lite$18/mo3× Claude Pro usage. ~80 prompts/5h, ~400/week. Includes GLM-4.7
Pro$72/mo5× Lite usage. ~400 prompts/5h, ~2,000/week. Includes GLM-5
Max$160/mo4× Pro usage. ~1,600 prompts/5h, ~8,000/week. Includes GLM-5.1
No separate Ultra tier
GLM-4.7GLM-4.5-AirGLM-5GLM-5.1
Free MCP: Vision, Web Search, Web Reader
KimiMoonshot AI
Moderato$19/moBasic Kimi Code usage + general membership benefits
Allegretto$39/mo5× more Kimi Code requests + general membership benefits
Allegro$99/mo15× more Kimi Code requests
Vivace$199/moHighest weekly plan quotas for complex projects
Kimi K2.5Kimi K2.6Kimi K2.5 (Reasoning)
100 TPS, 30 concurrent req, 256K context. Weekly quota refresh
ClaudeAnthropic
Pro$20/moIncludes Claude Code, Claude Cowork, unlimited projects, research access
Max 5X$100/mo5× more usage than Pro. Higher output limits, early access to new features
Max 20X$200/mo20× more usage than Pro. Priority access at peak times
No separate Ultra tier
Claude Opus 4.6Claude Opus 4.7Claude Opus 4.8Claude Sonnet 4.5Claude Sonnet 4.6Claude Haiku 4.5
Industry gold standard for reasoning. Claude Code included
MiniMaxMiniMax AI
Plus$20/mo~1.7B tokens/mo MiniMax-M3. 3–4 concurrent agents, 1M context, multimodal
Max$50/mo~5.1B tokens/mo MiniMax-M3. 4–5 concurrent agents, 1M context, video gen 3 clips/day
Ultra$120/mo~9.8B tokens/mo MiniMax-M3. 6–7 concurrent agents, 1M context, video gen 5 clips/day
No separate Scale tier
MiniMax-M2.5MiniMax-M2.7MiniMax-M3MiniMax-M2.5-HS
100 TPS. OpenClaw agent support. 10× weekly carryover
DevinCognition AI
Core$500/mo250 ACUs (Agentic Computing Units) included. $2/ACU overage. Autonomous software engineer agent
TeamCustomCustom ACU limits, dedicated onboarding, SSO, audit logging, enterprise knowledge graph
No separate Plus tier
No Scale tier
Devin (autonomous)
67% PR merge rate on defined tasks. Used by Citi, Mercedes, Goldman Sachs

Aggregators & Plugins

22
BRAND LITE / ENTRY PRO / STANDARD PLUS / MAX SCALE / ULTRA MODELS NOTES
Nous PortalNous Research
Free$0/moFree models only, $0 monthly credits
Plus$20/mo300+ models, hosted tool usage, $22 monthly credits, $10 rollover cap
Super$100/mo300+ models, hosted tool usage, $110 monthly credits, $50 rollover cap
Ultra$200/mo300+ models, hosted tool usage, $220 monthly credits, $100 rollover cap
300+ modelsClaudeGPTGeminiDeepSeekQwen
Unified access with hosted tools
OpenCode GoAnomaly
Go$5 first mo, then $10/mo14 open-source coding models. Rolling 5h limits: GLM-5.1 880 · Qwen3.7 Max 950 · Kimi K2.6 1,150 · MiniMax M3 1,400 · MiMo-V2.5-Pro 3,250 · Qwen3.6 Plus 3,300
Single tier
Single tier
Single tier
Kimi K2.6GLM-5.1Qwen3.6 PlusDeepSeek V4 ProMiniMax M2.7MiMo-V2.5-ProQwen3.7 MaxQwen3.5-Plus
MIT terminal agent. Works with Hermes, OpenClaw
Command CodeCredit-based
Go$1/mo$10 credits. taste-1 model, open-source models, ~15K requests
Pro$15/mo$30 credits. taste-1, open-source + premium models, ~25K requests, analytics
Max$100/mo$150 credits. taste-1, all models, ~110K requests, higher rate limits
Ultra$200/mo$300 credits. taste-1, all models, ~200K requests, highest rate limits
taste-1open-sourcepremium frontier
$1 = 10 credits
BytePlusAlibaba ModelArk
Lite$10/mo3× Claude Pro usage. ~30K requests/mo. 6 models included
Pro$50/mo5× Lite usage. ~90K requests/mo. All 6 models included
No Plus/Max tier
No Scale tier
Qwen3.5-PlusQwen3-MaxQwen3-Coder-NextQwen3-Coder-PlusGLM-4.7Kimi-K2.5
6 models under one subscription. Best multi-model flexibility
Alibaba CloudModel Studio
Lite$10/mo1,200 req/5h, 9,000/week, 18,000/month
Pro$50/mo6,000 req/5h, 45,000/week, 90,000/month
No higher tier
No Scale tier
Qwen3.5-PlusQwen3-MaxQwen3-Coder-NextQwen3-Coder-PlusGLM-4.7Kimi-K2.5
sk-sp- API key. China-based. Requires Alipay/UnionPay
Ollama CloudOllama.com
Free$0/moLight usage. Small models only
Pro$20/mo50× Starter usage. Run 3 cloud models simultaneously
Max$100/mo5× Pro usage. Priority access to new models
No higher tier
Llama 3.3Qwen3Qwen3.5DeepSeek-V4Gemma 2MistralMiMoGPT-OSS
Same `ollama run` CLI. Models run on their GPUs. :cloud suffix
CerebrasWafer-Scale
No free tier
Pro$50/mo24M tokens/day included. Pay-per-token overage
Max$200/mo120M tokens/day included. Highest throughput
No higher tier
GPT-OSS 120BLlama 3.1 405BDeepSeek V3.1Qwen3.5-397BGLM-4.7
Up to 3,000 tokens/sec. Wafer-scale inference
AI RouterSwiss-hosted
Unlimited$49/moUnlimited Qwen 3.6 27B. Swiss-hosted, no prompt logging. Fair-use policy
Single model plan
Single model plan
Single model plan
Qwen 3.6 27B
GDPR-compliant. No prompt logging. Swiss hosting
Hugging FaceHF Hub
Free tier$0/moFree tier available
Pro$9/mo20× inference usage. Access to private models, faster inference, no rate limits on public models
No higher tier
No Scale tier
500k+ HF Hub models
Hub access + faster inference. Models from your HF quota
Baidu CloudBaidu AI
Lite¥40/mo (~$5.50)18,000 req/mo
Pro¥200/mo (~$27.50)90,000 req/mo
No higher tier
No Scale tier
ERNIE-SpeedERNIE-LiteERNIE-4.5ERNIE-Code
China-based. Requires Alipay/UnionPay
Routing.shSwiss-hosted
Free$0/mo100 req/5h, DeepSeek-V3.2, Qwen3.5, basic support
Lite$10/mo150 req/5h, Kimi K2.6, MiniMax M2.7, email support
Premium$20/mo300 req/5h, preview models, priority TPS, priority support
Max$50/mo600 req/5h, full catalog, first access to preview models, dedicated support
DeepSeek-V3.2Qwen3.5Kimi K2.6MiniMax M2.7preview models
Model routing service with Swiss hosting
LLM GatewayUniversal gateway
Lite$29/moAll 200+ models, $87/mo in model usage (3× value). Casual hobby work
Pro$79/moAll 200+ models, $237/mo in model usage (3× value). Full-day agent runs
Max$179/moAll 200+ models, $537/mo in model usage (3× value). Non-stop heavy sessions
No separate Ultra tier
200+ models across all providers
Universal LLM API gateway with 3× value markup
Continue.devOpen Source
Starter$3/1M tokensPay-as-you-go input + output tokens. Bring your own API keys. MIT licensed. VS Code, JetBrains, Neovim
Team$20/seat/moCentralized management, shared context, team billing, organization-wide model selection
No Plus/Max tier
No Scale tier
Any (BYO keys)
Open source. AI checks on every PR via GitHub status checks
CodeGPTModel-agnostic
Free$0/moOpen source core. Bring your own API keys. VS Code & JetBrains
TeamsCustomShared context, team history, centralized billing, enterprise features
No Plus/Max tier
No Scale tier
Your choice (BYO API key)
Thinks, plans, then builds. Studies codebase before coding
SupermavenFast copilot
Free$0/moFast code suggestions, 7-day data retention
Pro$10/mo1M token context window, adapts to your coding style, $5/month in Supermaven Chat credits
No Plus/Max tier
No Scale tier
Supermaven ownClaudeGPT
Fastest copilot. 1M token context on Pro
Factory (Droid)Agent-native
Pro$20/moAgent-native CLI/IDE/SDK. Cloud & local background agents. Usage statistics
Plus$100/mo~5× Pro usage limits. Droid Computers: Factory-managed cloud VMs for remote agents
Max$200/mo~10× Pro usage limits. Early access to new features
TeamsCustomUp to 150 seats. SSO, SAML/SCIM, Zero Data Retention, admin controls
GPT-5Claude Opus/SonnetGeminiopen-weight
One subscription gives access to both Anthropic and OpenAI models
Kilo PassKilo Code
Starter$19/moUp to $26.60/mo in credits
Pro$49/moUp to $68.60/mo in credits
Expert$199/moUp to $278.60/mo in credits
No higher tier
multiple coding-optimized
Credit-based system for Kilo Code platform

IDE & Tool Bundles

14
BRAND FREE / ENTRY PRO / STANDARD PLUS / MAX SCALE / ULTRA MODELS NOTES
GitHub CopilotUsage-based
Free$0/mo2,000 completions/month, 50 chats/month, limited premium requests
Pro$10/mo300 premium requests/month, unlimited completions, Copilot coding agent, code review
Pro+$39/mo1,500 premium requests/month, access to all models including Claude Opus, GPT-5.4
Enterprise$39/user/mo1,000 premium requests/user/mo, admin controls, SSO, audit logs
GPT-5.4GPT-5.1Claude Opus 4.6Claude Sonnet 4.5Gemini 3.1 Proo3o3-mini
Usage-based from June 2026. 1 AI credit = $0.01
CursorAI Native IDE
HobbyFreeLimited Agent requests, limited Tab completions
Pro$20/moExtended Agent limits, frontier models, MCPs, cloud agents
Pro+$60/mo3× usage on all OpenAI, Claude, Gemini models
Ultra$200/mo20× usage, priority access to new features
GPT-5.4Claude Opus 4.6Gemini 3.1 ProCursor own
AI-native IDE. Most popular for professional dev
WindsurfCodeium
Free$0/mo25 credits/month, basic autocomplete
Pro$15/moUnlimited completions, 100 Agent tasks/mo, Cascade agent, MCPs
Teams$30/user/moAdmin controls, SSO, usage analytics, priority support
EnterpriseCustomCustom limits, on-prem deployment, dedicated support
Claude Sonnet 4.5GPT-4.1Gemini 2.5 ProCodeLlama
Cascade agent for autonomous coding. MCPs included
Amazon QAWS
Free$0/moBasic code completion, security scanning
Pro$19/user/mo50 agent requests/mo, chat, code generation, inline completion
No higher tier
No Scale tier
Amazon Q (proprietary)
AWS-native. Deep integration with AWS services
JetBrains AIJetBrains
No free tier
Pro€10/mo (~$11)AI code completion, chat, documentation generation for all JetBrains IDEs
Ultimate€30/mo (~$32)Full AI features + JetBrains All Products Pack access
Commercial€60/mo (~$65)Per-seat commercial licensing
JetBrains AIGPT-4oClaude Sonnet 3.5
Works in IntelliJ, PyCharm, WebStorm, etc.
TabninePrivacy-first
Starter$0/moBasic autocomplete, limited context
Pro$12/moFull code completion, custom model training, team context
Enterprise$39/user/moAir-gapped deployment, private model fine-tuning, zero data retention option
No Ultra tier
Tabnine ownClaude 3.5 SonnetGPT-4oCodeLlama (local)
Only major tool with fully air-gapped deployment
Augment CodeEnterprise context
No free tier
Indie$20/mo1,000 agentic requests/mo, context engine, codebase indexing
Standard$60/user/moUnlimited requests, advanced context, multi-repo support
Max$200/user/moPriority processing, custom integrations, dedicated support
Augment proprietaryGPT-4oClaude Sonnet 3.5DeepSeek V2.5
Enterprise-focused with deep codebase context engine
v0.devVercel UI gen
Free$0/mo~10 generations/month, public projects only
Pro$20/mo~200 generations/month, private projects, custom domains
Team$40/user/moUnlimited generations, team collaboration, SSO
No higher tier
Vercel v0 modelsReact/Next.js optimized
Rebranded v0.dev → v0.app Jan 2026. UI component generator
Bolt.newStackBlitz
Free$0/mo~10 projects/month, basic compute
Pro$20/moUnlimited projects, faster builds, private repos
Team$40/user/moTeam features, SSO, priority build queue
No higher tier
Bolt proprietary AIWebContainer-based
Browser-based full-stack app builder
LovableFull-stack builder
Starter$15/mo~100 builds/month, basic support
Pro$30/moUnlimited builds, priority support, custom domains
Team$60/user/moTeam collaboration, SSO, API access
No higher tier
Lovable proprietary AI
$100M ARR in 8 months. Fastest-growing startup ever
ReplitCloud IDE + Agent
Free$0/moPublic repls only, basic compute
Core$10/moPrivate repls, faster builds, 3 agents
Pro$25/moUnlimited private repls, 10 agents, priority support
Teams$40/user/moTeam management, SSO, admin controls
Replit AIexternal LLM API
Revenue grew 10× in 9 months after launching Agent
ClineOpen Source
Free$0/moOpen source. Bring your own API keys. VS Code extension
Teams$20/user/moTeam billing, centralized management, shared context
EnterpriseCustomCustom contracts, on-prem deployment, dedicated support
No Scale tier
ClaudeGPTGeminiDeepSeekQwenany OpenAI-compatible
VS Code extension. BYOK — bring any model. 5M+ installs
AiderTerminal-first
Open Source$0/moFree. Bring your own API keys. Runs in terminal
Aider Cloud$10–$30/moManaged hosting, includes model access
No higher tier
No Scale tier
GPT-4oClaude 3.5 SonnetDeepSeek V2.5Qwen3Llama 3any OpenAI-compatible
Terminal pair programmer. Git-native workflow
Kilo CodeVS Code agent
Free$0/mo100 requests/day, open-source models only
Pro$20/moUnlimited requests, premium models, Kilo Agent
Team$40/user/moTeam features, SSO, priority support
No higher tier
Qwen3.6 PlusDeepSeek V4 ProClaude Sonnet 4.5GPT-4.1GPT-4o
VS Code extension with autonomous agent mode
TraeAI-Native IDE
Free$0/moFull AI-native IDE, BYO API keys, agent mode, multi-file context
Pro$15/moManaged model access, priority support, team features
No Plus/Max tier
No Scale tier
Multiple (BYO keys)
Free full IDE. Open source. Agent mode included
ZedHigh-performance editor
Free$0/moOpen source core. AI features via bring-your-own API keys
Pro$20/moManaged AI access, higher rate limits, priority support
No Plus/Max tier
No Scale tier
ClaudeGPTopen-source
Rust-based editor. Extremely fast. GPGPU rendering