← Model list
GPT-5.2 Codex
openai·openai/gpt-5.2-codex·GA·Closed·gpt-codex series
Code-specialist GPT for repository edits, reviews, and long-running software agents
At a glance
Reference price$2.19 / $17.50 per 1M
Cache read $0.219 · Venice AI
Lowest paid$0.14 / $1.14
QiHang Gateway · 15.6× spread
Context256,000
⚠ Providers report 256,000–400,000; the table below is authoritative
Output limit65,536
Capabilities
✓ Reasoning✓ Tool use✓ Structured outputTemperature✓ Attachments
⚠ Providers report capability flags inconsistently
Modalities
TextImagePDF
Knowledge cutoff2025-08-31
Released / updated2025-12-11 / 2025-12-11
Quality & performanceArtificial Analysis · Intelligence Index v4.1 · rep. tier xhigh
Intelligence41.2
Value106
Coding—
Agentic—
Output speed—
TTFT—
Value formulaIQ 41.2 ÷ min blended $0.390 = 106
Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.
Available at 21 providers21 with public prices
| Provider | Tier | Input | Output | Cache read | Cache write | Context | Output limit | Status |
|---|---|---|---|---|---|---|---|---|
| QiHang | Gateway | $0.14 | $1.14 | — | — | 400,000 ⚠ | 128,000 | |
| Poe | Gateway | $1.60 | $13.00 | $0.16 | — | 400,000 ⚠ | 128,000 | |
| OrcaRouter | Gateway | $1.75 | $14.00 | $0.175 | — | 400,000 ⚠ | 128,000 | |
| NanoGPT | Gateway | $1.75 | $14.00 | $0.175 | — | 400,000 ⚠ | 128,000 | |
| Impossibl | Gateway | $1.75 | $14.00 | $0.175 | — | 400,000 ⚠ | 128,000 | |
| Jiekou.AI | Gateway | $1.75 | $14.00 | — | — | 400,000 ⚠ | 128,000 | |
| OpenRouter | Gateway | $1.75 | $14.00 | $0.175 | — | 400,000 ⚠ | 128,000 | |
| Vivgrid | Gateway | $1.75 | $14.00 | $0.175 | — | 400,000 ⚠ | 128,000 | |
| ZenMux | Gateway | $1.75 | $14.00 | $0.17 | — | 400,000 ⚠ | 64,000 | |
| Ofox | Gateway | $1.75 | $14.00 | $0.18 | — | 400,000 ⚠ | 128,000 | |
Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.
Your usage cost
1QiHang$85.00
2Poe$797.20
3ZenMux$860.40
4OrcaRouter$861.00
5NanoGPT$861.00
6Impossibl$861.00
The cheapest paid channel is the only channel.
Benchmark1 items
| Name | Conditions | Score | Metric | Source |
|---|---|---|---|---|
| SWE-Bench Pro | dataset: public | 41.04 | resolve rate | Source ↗ |
The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.
Reasoning control
effort = loweffort = mediumeffort = higheffort = xhigheffort = noneeffort = minimal
6 / 21 providers expose no reasoning control (reasoning_options: []).
Related models
GPT-5.3 Codexsame series$1.75 / $14.00GPT-5.3 Codex (OpenAI)same series$1.75 / $14.00GPT-5.3 Codex (Azure)same series$1.75 / $14.00GLM-5.2cheaper alternative$1.40 / $4.40GPT OSS 120Bcheaper alternative$1.00 / $4.20Kimi K2.6cheaper alternative$0.95 / $4.00DeepSeek V4 Procheaper alternative$0.435 / $0.87
Price historyone sample accumulated per data sync
Input listOutput listMin blended