DevPass
Your all-access pass to AI coding

Stop counting tokens.
Start shipping code.

One flat-rate subscription for Claude Code, Cursor, Cline, and every OpenAI-compatible tool. 200+ models, one API key, zero surprises on your bill.

Get your DevPassView plans
terminal
$ export ANTHROPIC_BASE_URL=https://api.llmgateway.io
$ export ANTHROPIC_AUTH_TOKEN=llmgtwy_your_key
$ claude
# works with any model — switch freely
$ export ANTHROPIC_MODEL=gpt-5
Works with
+ any OpenAI-compatible tool

Why developers switch to DevPass

Stop paying per token. Start shipping.

Predictable pricing

One flat monthly fee. No surprise bills or token counting. Just code with any model you want.

200+ models, one key

Claude, GPT-5, Gemini, Llama, Qwen, and every major model. Switch between them with an env var.

Resets every month

Your usage allowance refreshes automatically. No rollover anxiety, no manual top-ups.

2-minute setup

Set two environment variables and you're in. No SDK changes, no code refactoring.

Full observability

Track every request, session, and dollar spent. Real-time dashboards with cost and latency insights.

Upgrade anytime

Move between Lite, Pro, and Max as your needs change. No lock-in, cancel anytime.

Simple, transparent pricing

All plans include every model. Pick the usage level that fits your workflow.

Lite

For occasional AI-assisted coding

$29/mo
  • All 200+ models included
  • Usage resets monthly
Get started
Most popular

Pro

For daily development workflows

$79/mo
  • All 200+ models included
  • Usage resets monthly
  • Best value for developers
Get started

Max

For power users and heavy sessions

$179/mo
  • All 200+ models included
  • Usage resets monthly
  • Maximum throughput
Get started

Up and running in minutes

1

Pick a plan

Choose Lite, Pro, or Max. You get an API key immediately after subscribing.

2

Set your env vars

Point your tool's base URL to api.llmgateway.io and paste your key. Two lines, done.

3

Code with any model

Use Claude for architecture, GPT-5 for a second opinion, Gemini for speed — switch anytime.

Top coding models

All included with every plan — use whichever fits the task.

Recommended Coding Models

High-performance models optimized for coding tasks with tool support and prompt caching.

Claude Opus 4.5
claude-opus-4-5-20251101

Context: 200K

$5.00 in/$25.00 out/M tokens

Gemini 3 Pro (Preview)
gemini-3-pro-preview

Context: 1.0M

$2.00 in/$12.00 out/M tokens

Grok Code Fast 1
grok-code-fast-1

Context: 256K

$0.20 in/$1.50 out/M tokens

Grok 4.1 Fast Reasoning
grok-4-1-fast-reasoning

Context: 2.0M

$0.20 in/$0.50 out/M tokens

MiniMax M2.1
minimax-m2.1

Context: 197K

$0.27 in/$1.10 out/M tokens

Qwen3 Coder
qwen3-coder

Context: 262K

$0.22 in/$0.95 out/M tokens

GLM-4.7
glm-4.7

Context: 200K

$0.60 in/$2.20 out/M tokens

View all coding models

Stop watching your token balance

Pick a plan, set two env vars, and get back to building.

Get your DevPassBrowse models