CODEX SPECIALIST400K CONTEXTTEXT + VISIONPROMPT CACHINGPRIORITY AVAILABLE
GPT-5.3-Codex API Pricing
OpenAI's coding-specialized agent model: $1.75/M input, $14/M output, and $0.175/M cached input. GPT-5.3-Codex is optimized for long-running coding tasks in Codex-like environments and supports low through xhigh reasoning effort settings.
Input - per 1M tokens
$1.75/M
Codex specialist row standard
Output - per 1M tokens
$14.00/M
Priority output is $28/M standard
Cached input - 90% off
$0.17/M
Prompt cache hit price -90%
Effective - agentic blend
$1.54/M
92/8 split - 82% cache
§ 01 / TERMINAL
Run the numbers.
Live calculator pre-loaded with current GPT-5.3-Codex rates. Use it for long-running code agents, review loops, and repo-scale software work.
$ /mo
Workload split
Prompt cache hit rate
Tokens you can process
—
Words equivalent (English)
—
Effective rate
—
§ 02 / SCENARIOS
Real-world presets.
CODING AGENT
Feature branch implementation
$0.149/task
REFACTOR
Repo-scale refactor
$0.363/run
CODE REVIEW
Pull request review
$0.074/review
DEBUG LOOP
Failing test repair
$0.195/loop
PLANNING
Technical spec draft
$0.112/spec
§ 03 / TOKENIZER
Paste text. See tokens. See cost.
Exact · o200k_base Auto-counts as you type
Counted with the o200k_base BPE encoding — the same tokenizer the API uses — entirely in your browser. The encoder loads on your first keystroke (~1 MB, one time); your text never leaves this page.
Characters 583
Words 82
Tokens (exact) 144 tokens
Cost as input · uncached $0.000252 USD
Cost as output · uncached $0.002016 USD
Cost as cached input $0.000025 USD
| Model | Input /M | Output /M | Effective blended | Context | Best for |
|---|---|---|---|---|---|
| GPT-5.3-Codex Current | $1.75 cache $0.17 | $14.00 | $1.54 agentic 92/8 | 400K | Agentic coding tasks |
| GPT-5.4 | $2.50 cache $0.25 | $15.00 | $1.80 broader model | 1.05M | Affordable OpenAI frontier work |
| GPT-5.4 mini | $0.75 cache $0.07 | $4.50 | $0.54 cheaper subagent | 400K | Subagents and computer-use tasks |
| GPT-5.5 | $5.00 cache $0.50 | $30.00 | $3.61 flagship tier | 1M | Frontier coding and professional agents |
| Claude Opus 4.7 | $5.00 cache $0.50 | $25.00 | $3.21 Anthropic frontier | 1M | Anthropic frontier reasoning |
| Gemini 2.5 Pro | $1.25 cache $0.13 | $10.00 | $1.10 tier 1 pricing | 2M | Large multimodal context |
| DeepSeek V4 Pro | $0.43 cache $0.00 | $0.87 | $0.14 promo price | 1M | Low-cost reasoning workloads |
Frequently asked.
Practical GPT-5.3-Codex pricing questions, with OpenAI's published rates separated from workload assumptions.
Q · 01 What is the standard GPT-5.3-Codex API price? +
OpenAI lists
gpt-5.3-codex at $1.75/M input tokens, $0.175/M cached input tokens, and $14/M output tokens. This page stores the public list price in USD and marks the source as OpenAI's pricing docs.Q · 02 Is GPT-5.3-Codex still useful after GPT-5.4? +
Yes, but the positioning changed. GPT-5.4 is broader and slightly more expensive at about
$1.80/M blended, while GPT-5.3-Codex is a coding specialist at about $1.54/M. Use Codex when the workload is mostly software development and GPT-5.4 when broader professional work or computer use matters more.Q · 03 Is Batch pricing available for GPT-5.3-Codex? +
OpenAI's specialized-model pricing section does not list a Batch row for
gpt-5.3-codex. It lists Standard pricing and Priority pricing for Codex. This page therefore does not publish a Batch discount for GPT-5.3-Codex.Q · 04 What does Priority processing cost? +
The Priority table lists
gpt-5.3-codex at $3.50/M input, $0.35/M cached input, and $28/M output. That is 2x the standard price and should be reserved for latency-sensitive coding workflows.Q · 05 Does GPT-5.3-Codex have regional pricing? +
OpenAI's regional-processing note on the pricing page is attached to the GPT-5.4/5.5 family rows, not the specialized Codex row. This page therefore uses the global public Codex rate and does not publish a regional uplift for GPT-5.3-Codex.
Q · 06 What context window does GPT-5.3-Codex have? +
OpenAI's model page lists a
400,000 token context window and 128,000 max output tokens for GPT-5.3-Codex. The model supports text and image input with text output.Q · 07 What release date is used here? +
OpenAI introduced GPT-5.3-Codex on February 5, 2026. This page uses
2026-02-05 as the release date and keeps the price history flat unless OpenAI publishes a later price change.Q · 08 How accurate is the tokenizer estimate? +
The live widget uses an estimated
4.875 characters per token for English text and labels the tokenizer as tiktoken-o200k_base. It is good for budget planning, but exact billing can differ by language, whitespace, code, tool calls, and image inputs.