Codex + Claude Code · Usage-based pricing

AI coding models at 20% of the official reference price.

Compare input, output, and cache rates across every supported model. One clear 0.20× price ratio, shown in USD.

80% lower than reference21 model routes$10 · $50 · $100 credit packs coming next

Model catalog

Find the right route for your workflow

All displayed prices are Jaideepass rates per one million tokens. Original reference prices appear beside them.

Reference checked Jul 28, 2026

⌕

21 models

Claude CodeSave 80%

claude-fable-5

Claude-compatible coding route for general development, repository work, and longer agent sessions.

Input

$2/1MOriginal $10

Output

$10/1MOriginal $50

Cache read

$0.2/1MOriginal $1

Cache write

$2.5/1MOriginal $12.5

Token-basedAnthropic0.20× ratio

Claude CodeSave 80%

claude-haiku-4-5-20251001

A cost-efficient Claude Code option for quick edits, lightweight tasks, and frequent iterations.

Input

$0.2/1MOriginal $1

Output

$1/1MOriginal $5

Cache read

$0.02/1MOriginal $0.1

Cache write

$0.25/1MOriginal $1.25

Cache write · 1h

$0.4/1MOriginal $2

Dynamic pricingAnthropic0.20× ratio

Claude CodeSave 80%

claude-opus-4-5-20251101

A higher-capability Claude Code route for complex codebases, planning, and demanding reasoning tasks.

Input

$1/1MOriginal $5

Output

$5/1MOriginal $25

Cache read

$0.1/1MOriginal $0.5

Cache write

$1.25/1MOriginal $6.25

Cache write · 1h

$2/1MOriginal $10

Dynamic pricingAnthropic0.20× ratio

Claude CodeSave 80%

claude-opus-4-6

A higher-capability Claude Code route for complex codebases, planning, and demanding reasoning tasks.

Input

$1/1MOriginal $5

Output

$5/1MOriginal $25

Cache read

$0.1/1MOriginal $0.5

Cache write

$1.25/1MOriginal $6.25

Cache write · 1h

$2/1MOriginal $10

Dynamic pricingAnthropic0.20× ratio

Claude CodeSave 80%

claude-opus-4-7

A higher-capability Claude Code route for complex codebases, planning, and demanding reasoning tasks.

Input

$1/1MOriginal $5

Output

$5/1MOriginal $25

Cache read

$0.1/1MOriginal $0.5

Cache write

$1.25/1MOriginal $6.25

Cache write · 1h

$2/1MOriginal $10

Dynamic pricingAnthropic0.20× ratio

Claude CodeSave 80%

claude-opus-4-8

A higher-capability Claude Code route for complex codebases, planning, and demanding reasoning tasks.

Input

$1/1MOriginal $5

Output

$5/1MOriginal $25

Cache read

$0.1/1MOriginal $0.5

Cache write

$1.25/1MOriginal $6.25

Token-basedAnthropic0.20× ratio

Claude CodeSave 80%

claude-sonnet-4-5-20250929

Balanced Claude Code pricing for everyday coding, debugging, refactoring, and agent workflows.

Input

$0.6/1MOriginal $3

Output

$3/1MOriginal $15

Cache read

$0.06/1MOriginal $0.3

Cache write

$0.75/1MOriginal $3.75

Cache write · 1h

$1.2/1MOriginal $6

Dynamic pricingAnthropic0.20× ratio

Claude CodeSave 80%

claude-sonnet-4-6

Balanced Claude Code pricing for everyday coding, debugging, refactoring, and agent workflows.

Input

$0.6/1MOriginal $3

Output

$3/1MOriginal $15

Cache read

$0.06/1MOriginal $0.3

Cache write

$0.75/1MOriginal $3.75

Cache write · 1h

$1.2/1MOriginal $6

Dynamic pricingAnthropic0.20× ratio

Claude CodeSave 80%

claude-sonnet-5

Balanced Claude Code route for productive day-to-day development with lower token costs.

Input

$0.4/1MOriginal $2

Output

$2/1MOriginal $10

Cache read

$0.04/1MOriginal $0.2

Cache write

$0.5/1MOriginal $2.5

Cache write · 1h

$0.8/1MOriginal $4

Dynamic pricingAnthropic0.20× ratio

CodexSave 80%

codex-auto-review

Codex-compatible route focused on code review, repository checks, and automated feedback workflows.

Input

$1/1MOriginal $5

Output

$6/1MOriginal $30

Cache read

$0.1/1MOriginal $0.5

Token-basedOpenAI0.20× ratio

CodexSave 80%

gpt-5.3-codex-spark

A Codex-compatible route for fast coding tasks, iterative fixes, and repository assistance.

Input

$0.35/1MOriginal $1.75

Output

$2.8/1MOriginal $14

Cache read

$0.035/1MOriginal $0.175

Token-basedPartner route0.20× ratio

CodexSave 80%

gpt-5.3-codex-spark-openai-compact

Compact Codex-compatible route for fast coding tasks and tools expecting OpenAI-style requests.

Input

$0.35/1MOriginal $1.75

Output

$2.8/1MOriginal $14

Cache read

$0.035/1MOriginal $0.175

Token-basedPartner route0.20× ratio

CodexSave 80%

gpt-5.4

General Codex-compatible model route for coding, reasoning, tool use, and repository work.

Input

$0.5/1MOriginal $2.5

Output

$3/1MOriginal $15

Cache read

$0.05/1MOriginal $0.25

Token-basedOpenAI0.20× ratio

CodexSave 80%

gpt-5.4-mini

Lower-cost Codex-compatible model for quick changes, utility tasks, and high-volume workflows.

Input

$0.15/1MOriginal $0.75

Output

$0.9/1MOriginal $4.5

Cache read

$0.015/1MOriginal $0.075

Token-basedOpenAI0.20× ratio

CodexSave 80%

gpt-5.4-mini-openai-compact

Compact lower-cost route for quick coding tasks and OpenAI-compatible integrations.

Input

$0.15/1MOriginal $0.75

Output

$0.9/1MOriginal $4.5

Cache read

$0.015/1MOriginal $0.075

Token-basedOpenAI0.20× ratio

CodexSave 80%

gpt-5.4-openai-compact

Compact general-purpose route for coding agents and OpenAI-compatible development tools.

Input

$0.5/1MOriginal $2.5

Output

$3/1MOriginal $15

Cache read

$0.05/1MOriginal $0.25

Token-basedOpenAI0.20× ratio

CodexSave 80%

gpt-5.5

Higher-capability Codex-compatible route for difficult repository tasks and extended agent work.

Input

$1/1MOriginal $5

Output

$6/1MOriginal $30

Cache read

$0.1/1MOriginal $0.5

Token-basedOpenAI0.20× ratio

CodexSave 80%

gpt-5.5-openai-compact

Compact higher-capability route for coding agents and OpenAI-compatible development tools.

Input

$1/1MOriginal $5

Output

$6/1MOriginal $30

Cache read

$0.1/1MOriginal $0.5

Token-basedOpenAI0.20× ratio

CodexSave 80%

gpt-5.6-luna

Cost-efficient Codex-compatible route with separate pricing for standard and long-context work.

Input

$0.2/1MOriginal $1

Output

$1.2/1MOriginal $6

Cache read

$0.02/1MOriginal $0.1

Cache write

$0.25/1MOriginal $1.25

Dynamic pricingOpenAI0.20× ratio

CodexSave 80%

gpt-5.6-sol

Higher-capability Codex-compatible route with standard and long-context price tiers.

Input

$1/1MOriginal $5

Output

$6/1MOriginal $30

Cache read

$0.1/1MOriginal $0.5

Cache write

$1.25/1MOriginal $6.25

Dynamic pricingOpenAI0.20× ratio

CodexSave 80%

gpt-5.6-terra

Balanced Codex-compatible route with separate rates for standard and long-context workloads.

Input

$0.5/1MOriginal $2.5

Output

$3/1MOriginal $15

Cache read

$0.05/1MOriginal $0.25

Cache write

$0.625/1MOriginal $3.13

Dynamic pricingOpenAI0.20× ratio

Setup Guides

Connect in a few minutes

The addresses below are examples only. Your production Base URL, dashboard access, and API key are delivered after purchase.

Codex CLI

Add a custom model provider

Install and open Codex CLI.
Use the Base URL and API key from your delivery message.
Select a supported Codex model and run a test request.

Base URL: https://api.example.com/v1API Key: sk-your-api-keyModel: supported-codex-model

Claude Code

Set the gateway environment

Install and open Claude Code.
Set the delivered Base URL and authentication token.
Select a supported Claude model and run a test request.

ANTHROPIC_BASE_URL=https://api.example.comANTHROPIC_AUTH_TOKEN=sk-your-api-key

Pricing, documentation, and customer support are provided by Jaideepass. Dashboard access details and the production API Base URL are provided after purchase.

Model availability and reference pricing may change.