Compare / head-to-head
GPT-4ovs
Phi-4
GPT-4o leads 4 of 7 shared benchmarks. GPT-4o has the larger context window (128K tokens).
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
4 – 3
GPT-4o leads
Cheaper per token
—
list price, input + output
Larger context
GPT-4o
128K tokens
Providers
2 providers
OpenAI · Microsoft
GPT-4o
OpenAI · released 2024-05-13
textvision
Phi-4
Microsoft · released 2024-12-12
text
- Context
- 16K
- Max out
- 16K
- Input /1M
- —
- Output /1M
- —
- Cached /1M
- —
- Scores
- 17 · 7 core
Quality
Benchmark matrix
7 shared · 8 only GPT-4o · 6 only Phi-4| Benchmark | GPT-4o | Phi-4 | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 7 | ||||
| GPQA Diamond | 46% ↗ | 56.1% ↗ | -10.1 pt | Phi-4 |
| MMLU | 85.7% ↗ | 84.8% ↗ | +0.9 pt | GPT-4o |
| LMArena Elo | 1346.1 ↗ | 1216.6 ↗ | +129.5 | GPT-4o |
| BALROG (BALROG) | 32.3% ↗ | 11.6% ↗ | +20.7 pt | GPT-4o |
| Chess Puzzles (Epoch AI run) | 13% ↗epoch run | 1% ↗epoch run | +12 pt | GPT-4o |
| MATH Level 5 (Epoch AI run) | 53.3% ↗epoch run | 64.9% ↗epoch run | -11.6 pt | Phi-4 |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | 6.4% ↗epoch run | 13.8% ↗epoch run | -7.4 pt | Phi-4 |
| Only GPT-4o reports · 8 | ||||
| AIME 2025 | 11.7% ↗matharena | not reported | — | — |
| AIME 2024 | 13.1% ↗ | not reported | — | — |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | 0.4% ↗epoch run | not reported | — | — |
| GSO Opt@1 (GSO) | 0% ↗ | not reported | — | — |
| LMArena Vision (LMArena) | 1161.8 ↗ | not reported | — | — |
| SimpleQA Verified | 26% ↗epoch run | not reported | — | — |
| SWE-bench Verified (Epoch AI run) | 31% ↗epoch run | not reported | — | — |
| WeirdML (Håvard Tveit Ihle) | 25.1% ↗ | not reported | — | — |
| Only Phi-4 reports · 6 | ||||
| MMLU-Pro | not reported | 70.4% ↗ | — | — |
| HumanEval | not reported | 82.6% ↗ | — | — |
| ArenaHard | not reported | 75.4% ↗ | — | — |
| HumanEval+ | not reported | 82.8% ↗ | — | — |
| IFEval | not reported | 63% ↗ | — | — |
| MATH | not reported | 80.4% ↗ | — | — |
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is GPT-4o minus Phi-4 in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing
Side by side
Official pricing pages and model cards| Spec | GPT-4o | Phi-4 | Edge |
|---|---|---|---|
| Context window | 128K tokens | 16K tokens | GPT-4o |
| Max output | 16K tokens | 16K tokens | Tie |
| Input price / 1M | $2.5 ↗ | — | — |
| Output price / 1M | $10 ↗ | — | — |
| Cached input / 1M | $1.25 ↗ | — | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $12.5 | — | — |
| Modalities | text · vision | text | GPT-4o |
| Released | 2024-05-13 | 2024-12-12 | — |
| Cited benchmark scores | 21 | 17 | — |
Reliability
Provider status
Live from /statusMore matchups
GPT-4o vs …
Models sharing the most benchmarksMore matchups
Phi-4 vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run GPT-4o and Phi-4 on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.