GPT-5.3-Codex is OpenAI’s most advanced agentic coding model, combining the frontier software engineering performance of GPT-5.2-Codex with the broader reasoning and professional knowledge capabilities of GPT-5.2. It achieves state-of-the-art results...
Tested on 18 benchmarks · BenchGecko score 69.3. Top scores: Artificial Analysis · GPQA Diamond (91.5%), Artificial Analysis · tau2-Bench Telecom (86.0%), Artificial Analysis · Long Context Reasoning (83.3%).
Qwen3.5 397B A17B scores 69.3 (100% as good) at $0.60/1M input · 66% cheaper
Unusual and adversarial machine learning challenges. Tests robustness of reasoning about edge cases in ML systems.
Complex terminal-based engineering tasks. Models must use command-line tools, navigate filesystems, and debug systems through shell interaction.
Real-world software engineering tasks from GitHub issues. Models must diagnose bugs and write patches that pass test suites. Human-verified subset of SWE-bench.
Evaluates post-training behaviors including instruction following, safety, and helpfulness balance.
SEAL SWE Atlas Codebase Q&A. Tests understanding of large codebases through question answering.
Agent performance evaluation testing multi-step tool use, planning, and execution in realistic environments.
- Typemultimodal
- Context400K tokens (~200 books)
- ReleasedFeb 2026
- LicenseProprietary
- StatusActive
- Cost / Message~$0.018
Frequently Asked Questions
Related Models
Qwen3.5 397B A17B · Alibaba QwenClaude Opus 4.8 · Anthropicphi-3-medium 14B · MicrosoftGPT-5.2 Pro · OpenAIClaude Instant · AnthropicKey facts · as of 2026-11-11
- GPT-5.3-Codex by OpenAI. BenchGecko score 69.3, rank 44 of 326 scored models (normalized average of public benchmark scores).
- List price $1.75 input · $14.00 output per 1M tokens (as of 2026-11-11).
- Sold by 3 providers (as of 2026-11-11): Azure $1.75 in / $14.00 out · OpenAI $1.75 in / $14.00 out · OpenAI $3.50 in / $28.00 out. Every provider
How to cite · data as of 2026-11-11
GPT-5.3-Codex · benchmarks, pricing and providers. BenchGecko, data as of 2026-11-11. https://benchgecko.ai/model/gpt-5-3-codex
Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP