Very long inputs
Choose GPT-5.2-Codex
GPT-5.2-Codex reads up to 400k tokens per request, against 8k for Tiny Aya Global.
GPT-5.2-Codex vs Tiny Aya Global: price, context and capabilities
Choose GPT-5.2-Codex for very long inputs and images as input. Choose Tiny Aya Global for self-hosting and data control and the latest release.
In short
- A price comparison is not possible: one of them has no comparable published price.
- GPT-5.2-Codex reads more per request: 400k tokens against 8k.
- Tiny Aya Global is the newer model (Feb 17, 2026, against Jan 14, 2026).
- Only GPT-5.2-Codex offers vision, reasoning, tool calling, structured output, batch api.
- Only Tiny Aya Global offers self-hostable.
Which should you choose?
Each answer below comes straight from the published prices, specifications and independent benchmark results.
Self-hosting and data control
Choose Tiny Aya Global
Tiny Aya Global has open weights, so you can run it on your own servers; GPT-5.2-Codex is only available through its provider.
Images as input
Choose GPT-5.2-Codex
GPT-5.2-Codex accepts images as input; Tiny Aya Global does not.
Latest release
Choose Tiny Aya Global
Tiny Aya Global is the newer release (Feb 17, 2026, against Jan 14, 2026 for GPT-5.2-Codex).
Specifications side by side
| GPT-5.2-Codex | Tiny Aya Global | |
|---|---|---|
| Provider | OpenAI | Cohere |
| Category | Language models | Language models |
| Status | Deprecated | Generally available |
| Headline price | No data | No data |
| Input | No data | No data |
| Output | No data | No data |
| Cached input | No data | No data |
| Context window | 400k | 8k |
| Max output | 128k | No data |
| Released | Jan 14, 2026 | Feb 17, 2026 |
| Knowledge cutoff | August 2025 | No data |
| Licence | Proprietary | Open weights |
| Capabilities | Vision, Reasoning, Tool calling, Structured output, Batch API | Self-hostable |
| API model id | gpt-5.2-codex | tiny-aya-global |
Benchmarks side by side
Leader in bold.
| Benchmark | GPT-5.2-Codex | Tiny Aya Global |
|---|---|---|
| GPQA Diamond | 89.9% | 30.5% |
| Humanity's Last Exam | 35.7% | 5.2% |
| IFBench (instruction following) | 77.6% | 20.1% |
| Terminal-Bench Hard | 37.1% | No data |
| τ²-Bench (agentic tool use) | 92.1% | No data |
| MMMU-Pro (visual reasoning) | 76.3% | No data |
| Long-context reasoning (AA-LCR) | 82.3% | No data |
Independent evaluations by Artificial Analysis, run the same way for every model, at the reasoning setting shown. Percentages unless marked as an index.
Frequently asked questions
Should I choose GPT-5.2-Codex or Tiny Aya Global?
Choose GPT-5.2-Codex for very long inputs and images as input. Choose Tiny Aya Global for self-hosting and data control and the latest release.
Which is cheaper, GPT-5.2-Codex or Tiny Aya Global?
A price comparison is not possible: one of them has no comparable published price.
Which has the larger context window, GPT-5.2-Codex or Tiny Aya Global?
GPT-5.2-Codex reads more per request: 400k tokens against 8k.
What is the difference between GPT-5.2-Codex and Tiny Aya Global?
GPT-5.2-Codex is made by OpenAI and Tiny Aya Global by Cohere. A price comparison is not possible: one of them has no comparable published price. GPT-5.2-Codex reads more per request: 400k tokens against 8k. Tiny Aya Global is the newer model (Feb 17, 2026, against Jan 14, 2026). Only GPT-5.2-Codex offers vision, reasoning, tool calling, structured output, batch api. Only Tiny Aya Global offers self-hostable.
Which is newer, GPT-5.2-Codex or Tiny Aya Global?
Tiny Aya Global was released on Feb 17, 2026 and GPT-5.2-Codex on Jan 14, 2026.