Compare / head-to-head
Phi-4vs
Seed-OSS-36B-Instruct
Seed-OSS-36B-Instruct leads 6 of 6 shared benchmarks. Seed-OSS-36B-Instruct has the larger context window (512K tokens).
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
0 – 6
Seed-OSS-36B-Instruct leads
Cheaper per token
—
list price, input + output
Larger context
Seed-OSS-36B-Instruct
512K tokens
Providers
2 providers
Microsoft · ByteDance Seed
Phi-4
Microsoft · released 2024-12-12
text
- Context
- 16K
- Max out
- 16K
- Input /1M
- —
- Output /1M
- —
- Cached /1M
- —
- Scores
- 17 · 7 core
Seed-OSS-36B-Instruct
ByteDance Seed · released 2025-08-20
text
- Context
- 512K
- Max out
- —
- Input /1M
- —
- Output /1M
- —
- Cached /1M
- —
- Scores
- 18 · 12 core
Quality
Benchmark matrix
6 shared · 7 only Phi-4 · 12 only Seed-OSS-36B-Instruct| Benchmark | Phi-4 | Seed-OSS-36B-Instruct | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 6 | ||||
| GPQA Diamond | 56.1% ↗ | 71.5% ↗epoch run | -15.4 pt | Seed-OSS-36B-Instruct |
| MMLU | 84.8% ↗ | 87.4% ↗ | -2.6 pt | Seed-OSS-36B-Instruct |
| MMLU-Pro | 70.4% ↗ | 82.7% ↗ | -12.3 pt | Seed-OSS-36B-Instruct |
| Chess Puzzles (Epoch AI run) | 1% ↗epoch run | 13% ↗epoch run | -12 pt | Seed-OSS-36B-Instruct |
| IFEval | 63% ↗ | 85.8% ↗ | -22.8 pt | Seed-OSS-36B-Instruct |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | 13.8% ↗epoch run | 67.5% ↗epoch run | -53.7 pt | Seed-OSS-36B-Instruct |
| Only Phi-4 reports · 7 | ||||
| HumanEval | 82.6% ↗ | not reported | — | — |
| LMArena Elo | 1216.6 ↗ | not reported | — | — |
| ArenaHard | 75.4% ↗ | not reported | — | — |
| BALROG (BALROG) | 11.6% ↗ | not reported | — | — |
| HumanEval+ | 82.8% ↗ | not reported | — | — |
| MATH | 80.4% ↗ | not reported | — | — |
| MATH Level 5 (Epoch AI run) | 64.9% ↗epoch run | not reported | — | — |
| Only Seed-OSS-36B-Instruct reports · 12 | ||||
| AIME24 | not reported | 91.7% ↗ | — | — |
| AIME25 | not reported | 84.7% ↗ | — | — |
| BeyondAIME | not reported | 65% ↗ | — | — |
| GPQA-D | not reported | 71.4% ↗ | — | — |
| HLE | not reported | 10.1% ↗ | — | — |
| LiveCodeBench v6 (02/2025-05/2025) | not reported | 67.4% ↗ | — | — |
| MMMLU | not reported | 78.4% ↗ | — | — |
| Multi-SWE-Bench | not reported | 17% ↗ | — | — |
| RULER (128K) | not reported | 94.6% ↗ | — | — |
| SuperGPQA | not reported | 55.7% ↗ | — | — |
| SWE-Bench Verified (OpenHands) | not reported | 56% ↗ | — | — |
| TAU1-Retail | not reported | 70.4% ↗ | — | — |
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 Phi-4 minus Seed-OSS-36B-Instruct 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 | Phi-4 | Seed-OSS-36B-Instruct | Edge |
|---|---|---|---|
| Context window | 16K tokens | 512K tokens | Seed-OSS-36B-Instruct |
| Max output | 16K tokens | — | — |
| 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 | 2024-12-12 | 2025-08-20 | — |
| Cited benchmark scores | 17 | 18 | — |
Reliability
Provider status
Live from /statusMore matchups
Phi-4 vs …
Models sharing the most benchmarksMore matchups
Seed-OSS-36B-Instruct vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run Phi-4 and Seed-OSS-36B-Instruct on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.