DevPass
Built for Claude Code · OpenCode · SoulForge

One key. Every model.
Three flat prices.

DevPass turns every dollar you spend into $3 of model usage at provider rates. Pair it with SoulForge and prompt caching pushes that to roughly $6.

Works the same in Claude Code, OpenCode, SoulForge, Cline, and every OpenAI-compatible tool — no SDK changes.

Get your DevPassSee pricing
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

Native fit

Drop-in for the agents you already use

DevPass is built around how Claude Code, OpenCode, and SoulForge actually work — not a generic OpenAI-compatible proxy you have to glue together.

Claude Code

Two env vars and Claude Code routes through LLM Gateway. Use any model — Claude, GPT-5, Gemini, GLM — with a single ANTHROPIC_MODEL flip.

ANTHROPIC_BASE_URL + AUTH_TOKEN

OpenCode

LLM Gateway is built into OpenCode. Run `opencode`, type `/connect`, paste your DevPass key. No env vars, no config files.

/connect → LLM Gateway
SOULFORGE
Saves ~50% tokens

SoulForge

Aggressive prompt caching cuts roughly 50% of tokens on multi-turn agent runs. Pair with DevPass to effectively double your monthly usage.

/keys → paste your key
+ Cline, Cursor, Aider, Continue & any OpenAI-compatible tool
Pair with SoulForge

Cut ~50% of tokens.
Double the value of your DevPass.

SoulForge is a coding agent built around aggressive prompt caching and context reuse. Point it at LLM Gateway and it sends roughly half the tokens of an equivalent Claude Code session — same model, same task, smaller bill.

  • 1
    Prompt caching by default — system prompt, tools, and project context are cached on every provider that supports it.
  • 2
    Context-aware compaction — SoulForge prunes stale turns instead of replaying the whole conversation.
  • 3
    Same DevPass key — no separate subscription. Run soulforge, type /keys, paste your key.
Get SoulForge
Tokens used · same task
Without SoulForge1,000K

Standard agent loop, no aggressive caching

SOULFORGEWith SoulForge−50%
~500K

Prompt-cache hits on every reusable prefix

Effective DevPass value
3×
Plan baseline
~6×
With SoulForge

Actual savings vary by workload. The 50% figure is typical for multi-turn agent sessions where the system prompt and codebase context stay stable.

Pricing

What you pay vs. what you get

Every plan includes the full 200+ model catalog. The only thing that changes is the size of your monthly usage allowance.

Lite

For occasional AI-assisted coding

$29/mo
Casual hobby work
What you actually get3× value
You pay$29/mo
You use$87 at provider rates
$29 paid$87 used
With SoulForge → prompt caching cuts ~50% of tokens, stretching your $87 to ~$174 of effective use.
  • All 200+ models — Claude, GPT-5, Gemini, GLM, Qwen, …
  • Works with Claude Code, OpenCode, SoulForge & every OpenAI-compatible tool
  • Real-time usage dashboard with per-request cost
  • Switch tiers any time — prorated
Get Lite
Most popular

Pro

For daily development workflows

$79/mo
Most developers ship from here
What you actually get3× value
You pay$79/mo
You use$237 at provider rates
$79 paid$237 used
With SoulForge → prompt caching cuts ~50% of tokens, stretching your $237 to ~$474 of effective use.
  • Everything in Lite
  • Headroom for full-day agent runs in Claude Code & OpenCode
  • Priority routing on flagship models
  • Email support with 1-business-day reply
Get Pro

Max

For power users and heavy sessions

$179/mo
All-day agent runs
What you actually get3× value
You pay$179/mo
You use$537 at provider rates
$179 paid$537 used
With SoulForge → prompt caching cuts ~50% of tokens, stretching your $537 to ~$1074 of effective use.
  • Everything in Pro
  • Comfortable for non-stop SoulForge & Claude Code usage
  • Priority support, faster turnaround
  • Best $/usage ratio across the lineup
Get Max

Usage is metered at each provider's published per-token rate. Every request shows its dollar value in your dashboard in real time.

Compare every feature on the pricing page

Up and running in minutes

01

Pick a plan

Choose Lite, Pro, or Max. Your DevPass key works everywhere — no separate keys per tool.

02

Plug it into your agent

Two env vars for Claude Code, /providers in OpenCode, /keys in SoulForge. No SDK changes, no code refactor.

03

Switch models freely

Claude Opus 4.7 for architecture, GPT-5.5 for review, Gemini 3.1 Pro for fresh eyes — same key, no extra cost.

The latest flagships

Every plan ships with the newest models

Claude Opus 4.7, Gemini 3.1 Pro, GPT-5.5 Pro, plus the strongest open-weight Chinese coders — included on every tier.

Recommended Coding Models

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

GPT-5.5 Pro
gpt-5.5-pro

Context: 1.1M

$30.00 in/$180.00 out/M tokens

GPT-5.3 Codex
gpt-5.3-codex

Context: 400K

$1.75 in/$14.00 out/M tokens

Claude Opus 4.7
claude-opus-4-7

Context: 1.0M

$5.00 in/$25.00 out/M tokens

Gemini 3.1 Pro (Preview)
gemini-3.1-pro-preview

Context: 1.0M

$2.00 in/$12.00 out/M tokens

DeepSeek V4 Pro
deepseek-v4-pro

Context: 1.1M

$0.43 in/$0.87 out/M tokens

Kimi K2.6
kimi-k2.6

Context: 262K

$0.95 in/$4.00 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

FAQ

Common questions

Everything you need to know about usage limits, plans, and getting started with DevPass.

Can't find an answer? Contact us

Every plan includes a fixed monthly usage allowance, calculated at standard provider rates:

  • Lite — $29/mo: $87 in monthly model usage
  • Pro — $79/mo: $237 in monthly model usage
  • Max — $179/mo: $537 in monthly model usage

That's roughly 3× the subscription price — effectively model access at a deep discount compared to paying providers directly.

Stop watching your token balance

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

Get your DevPassBrowse all models