Compare / head-to-head
DeepSeek V3.2vsPhi-4 Reasoning
DeepSeek V3.2 leads 3 of 3 shared benchmarks. DeepSeek V3.2 has the larger context window (131K tokens).
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
3 – 0
DeepSeek V3.2 leads
Cheaper per token
—
list price, input + output
Larger context
DeepSeek V3.2
131K tokens
Provider uptime (30d)
100% · —
DeepSeek · Microsoft
DeepSeek V3.2
DeepSeek · released 2025-12-01
text
- Context
- 131K
- Max out
- 66K
- Input /1M
- —
- Output /1M
- —
- Cached /1M
- —
- Scores
- 8 · 7 core
Phi-4 Reasoning
Microsoft · released 2025-04-30
text
- Context
- 33K
- Max out
- 33K
- Input /1M
- —
- Output /1M
- —
- Cached /1M
- —
- Scores
- 4 · 4 core
Quality
Benchmark matrix
3 shared · 4 only DeepSeek V3.2 · 1 only Phi-4 Reasoning| Benchmark | DeepSeek V3.2 | Phi-4 Reasoning | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 3 | ||||
| GPQA Diamond | 82.4% ↗ | 65.8% ↗ | +16.6 pt | DeepSeek V3.2 |
| MMLU-Pro | 85% ↗ | 74.3% ↗ | +10.7 pt | DeepSeek V3.2 |
| AIME 2025 | 93.1% ↗ | 62.9% ↗ | +30.2 pt | DeepSeek V3.2 |
| Only DeepSeek V3.2 reports · 4 | ||||
| SWE-bench Verified | 73.1% ↗ | not reported | — | — |
| AIME 2026 | 94.2% ↗matharena | not reported | — | — |
| Humanity's Last Exam (no tools) | 25.1% ↗ | not reported | — | — |
| LMArena Elo | 1424.8 ↗ | not reported | — | — |
| Only Phi-4 Reasoning reports · 1 | ||||
| HumanEval+ | not reported | 92.9% ↗ | — | — |
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 DeepSeek V3.2 minus Phi-4 Reasoning 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 | DeepSeek V3.2 | Phi-4 Reasoning | Edge |
|---|---|---|---|
| Context window | 131K tokens | 33K tokens | DeepSeek V3.2 |
| Max output | 66K tokens | 33K tokens | DeepSeek V3.2 |
| Input price / 1M | — | — | — |
| Output price / 1M | — | — | — |
| Cached input / 1M | — | — | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | — | — | — |
| Modalities | text | text | Tie |
| Released | 2025-12-01 | 2025-04-30 | — |
| Cited benchmark scores | 8 | 4 | — |
Reliability
Provider status
Last 60 days · refreshed daily on this page · live on /statusMore matchups
DeepSeek V3.2 vs …
Models sharing the most benchmarksMore matchups
Phi-4 Reasoning vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run DeepSeek V3.2 and Phi-4 Reasoning on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.