OpenAI model product
GPT-5.3-Codex
3-Codex is OpenAI's most capable coding model, combining frontier agentic coding capabilities, improvements in aesthetics, and context compaction.
Updated Aug 12, 2026. Default version: GPT-5.3 Codex
LLMBoard score62.9GPT-5.3 Codex
Coverage20%6 benchmark families
Context window400KTokens
Official input price$1.75OpenAI API
GPT-5.3-Codex Specifications
Technical details for the model's default version.
- Version
- GPT-5.3 Codex
- Released
- Feb 5, 2026
- Knowledge cutoff
- Aug 31, 2025
- Parameters
- N/A
- Context window
- 400K
- Max output
- 128K
- Inputs
- image, pdf, text
- Outputs
- text
- Open weights
- No
- License
- Proprietary
GPT-5.3-Codex Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
GPT-5.3-Codex Benchmark Results
Benchmark scores for GPT-5.3 Codex.
| Cybersecurity CTFs | 77.6% | 01 | 3 | 100.0% | C | |
| SWE-Lancer (IC-Diamond subset) | 81.4% | 02 | 6 | 80.0% | C | |
| Terminal-Bench 2.0 | 77.3% | 04 | 49 | 93.8% | C | |
| OSWorld-Verified | 64.7% | 18 | 23 | 22.7% | C | |
| LiveBench | 72.8% | 23 | 38 | 40.5% | B | |
| SWE-Bench Pro | 56.8% | 25 | 45 | 45.5% | C |
GPT-5.3-Codex Arena Results
Preference and agent-evaluation results for the default version.
| webdev | overall | 56 | 1408.5 | 2,524 | N/A | |
| webdev | overall | 72 | 1369.9 | 3,493 | N/A |
GPT-5.3-Codex Runtime Performance
Provider-specific output speed and catalog latency for GPT-5.3 Codex. Runtime does not affect the capability score.
| OpenAI | 2.426 tok/s | 7.286 s | 400K | 128K |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
GPT-5.3-Codex Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $1.75 input, $14 output per 1M
- Official provider
- OpenAI
- Lowest third-party
- From $1.6 input, $13 output per 1M via Poe
- Tracked offerings
- 24
| Poe | openai/gpt-5.3-codex | global | $1.6 | $13 | 400K | |
| Impossibl | openai/gpt-5.3-codex | global | $1.75 | $14 | 400K | |
| Vivgrid | gpt-5.3-codex | global | $1.75 | $14 | 400K | |
| NanoGPT | openai/gpt-5.3-codex | global | $1.75 | $14 | 400K | |
| FastRouter | openai/gpt-5.3-codex | global | $1.75 | $14 | 400K | |
| LLM Gateway | gpt-5.3-codex | global | $1.75 | $14 | 400K | |
| OpenCode Zen | gpt-5.3-codex | global | $1.75 | $14 | 400K | |
| GitHub Copilot | gpt-5.3-codex | global | $1.75 | $14 | 400K | |
| Neon | gpt-5-3-codex | global | $1.75 | $14 | 400K | |
| FrogBot | gpt-5-3-codex | global | $1.75 | $14 | 400K | |
| Pioneer | gpt-5.3-codex | global | $1.75 | $14 | 400K | |
| ZenMux | openai/gpt-5.3-codex | global | $1.75 | $14 | 400K | |
| Abacus | gpt-5.3-codex | global | $1.75 | $14 | 400K | |
| Azure Cognitive Services | gpt-5.3-codex | global | $1.75 | $14 | 400K | |
| Azure | gpt-5.3-codex | global | $1.75 | $14 | 400K | |
| AIHubMix | gpt-5.3-codex | global | $1.75 | $14 | 400K | |
| OrcaRouter | openai/gpt-5.3-codex | global | $1.75 | $14 | 400K | |
| FreeModel | gpt-5.3-codex | global | $1.75 | $14 | 400K | |
| Kilo Gateway | openai/gpt-5.3-codex | global | $1.75 | $14 | 400K | |
| Vercel AI Gateway | openai/gpt-5.3-codex | global | $1.75 | $14 | 400K | |
| OpenRouter | openai/gpt-5.3-codex | global | $1.75 | $14 | 400K | |
| Cloudflare AI Gateway | openai/gpt-5.3-codex | global | $1.75 | $14 | 400K | |
| OpenAI | gpt-5.3-codex | global | $1.75 | $14 | 400K | |
| Venice AI | openai-gpt-53-codex | global | $2.19 | $17.5 | 400K |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
GPT-5.3-Codex Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
| GPT-5.3 Codex | 62.9 | N/A | 400K | 128K | No | Proprietary |
What is GPT-5.3-Codex?
Key information about GPT-5.3-Codex and its available data.
3-Codex is OpenAI's most capable coding model, combining frontier agentic coding capabilities, improvements in aesthetics, and context compaction. 4%). First model classified as High capability for cybersecurity under OpenAI's Preparedness Framework. Available in the Codex app and API.
Data as of 2026-08-11.
GPT-5.3-Codex vs nearby models
Open a comparison with the three ranked models immediately above and below this model.
Models similar to GPT-5.3-Codex
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
OP
GPT-5.1
OpenAI
OP
GPT-5.1-Instant
OpenAI
OP
GPT-5.4-mini
OpenAI
OP
GPT-5.2-Codex
OpenAI
OP
GPT-5.2
OpenAI
OP
GPT-5.2-Pro
OpenAI
FAQ
Common questions about GPT-5.3-Codex.
When was GPT-5.3-Codex released?
GPT-5.3-Codex's default version was released on Feb 5, 2026.
How much does GPT-5.3-Codex cost?
GPT-5.3-Codex's official API price is $1.75 per million input tokens and $14 per million output tokens via OpenAI. The lowest tracked third-party offer starts at $1.6 input and $13 output via Poe.
Who created GPT-5.3-Codex?
GPT-5.3-Codex was created by OpenAI.
What is the context window for GPT-5.3-Codex?
The default version has a 400K token context window.
Is GPT-5.3-Codex open weight?
No. The default version is not marked as having publicly available weights.
How many API providers offer GPT-5.3-Codex?
24 provider offerings are linked to the default version.
What models should I compare GPT-5.3-Codex with?
Nearby ranked alternatives include Kimi K2.5, Gemini 3 Flash, MiMo V2 Pro.