Models · Compare
Phi-4-mini Instruct vs GPT-4.5
Rows highlighted in warm gray are where the models differ. Numbers carry their as-of date and primary source.
Specs
| Field | A: Phi-4-mini Instruct | B: GPT-4.5 |
|---|---|---|
| Released | — | 2025-02-27 |
| Developer | Microsoft | OpenAI |
| Openness | Open | Proprietary |
| License | MIT | Proprietary |
| OSI-approved | yes | no |
| Data released | no | no |
| Training code | no | no |
| Architecture | dense | unknown |
| Total params | — | — |
| Active params | — | — |
| Experts | — | — |
| Context window | 128K | 128K |
| Attention | gqa | unknown |
| Position enc. | rope | unknown |
| Pretraining tokens | — | — |
| Post-training | sft, dpo | rlhf |
| Training hardware | A100 | — |
| $/M input | — | $0.00 |
| $/M output | — | $0.00 |
| Output tok/sec | — | 0 |
Benchmarks
Missing scores render as not reported; never inferred. Bold highlights the leader per benchmark.
General reasoning
| MMLU | 67.3 2025-02-26 | — |
Math
| MATH | 64.0 2025-02-26 | — |
Context · A
Small-tier Phi 4 released February 2025: 3.8B dense decoder-only with 128K context, 200K vocab, and grouped-query attention. Trained on 5T tokens for 21 days on 512 A100-80G GPUs, with a data cutoff of June 2024. Supports 22 languages.
Context · B
Released February 27, 2025 as a research preview of OpenAI's largest pretrained chat model, internally referred to as Orion. Sam Altman characterized it as a giant, expensive model; pricing came in at $75 input and $150 output per million tokens, roughly 30x GPT-4o's input rate at launch.