ModelsCompareBest forBenchmarksStatusPricingAPI
Compare / head-to-head

MAI-Thinking-1vsPhi-4 Reasoning

MAI-Thinking-1 leads 3 of 3 shared benchmarks. MAI-Thinking-1 has the larger context window (256K tokens).

Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.

Shared benchmarks
3 – 0
MAI-Thinking-1 leads
Cheaper per token
list price, input + output
Larger context
MAI-Thinking-1
256K tokens
Provider uptime (30d)
Microsoft
MAI-Thinking-1
Microsoft · released 2026-08-12
text
Context
256K
Max out
64K
Input /1M
Output /1M
Cached /1M
Scores
15 · 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

BenchmarkMAI-Thinking-1Phi-4 ReasoningΔEdge
Reported by both · 3
GPQA Diamond84.2% 65.8% +18.4 ptMAI-Thinking-1
MMLU-Pro85% 74.3% +10.7 ptMAI-Thinking-1
AIME 202597% 62.9% +34.1 ptMAI-Thinking-1
Only MAI-Thinking-1 reports · 12
SWE-bench Verified73.5% not reported
AIME 202694.5% not reported
AdvancedIF85% not reported
BFCL v372% not reported
GraphWalks (<=128K)90% not reported
HMMT February 202684.9% not reported
IFBench69% not reported
LiveCodeBench v687.7% not reported
MultiChallenge53% not reported
SimpleQA Verified31% not reported
SWE-bench Pro52.8% not reported
Terminal-Bench 2.046% not reported
Only Phi-4 Reasoning reports · 1
HumanEval+not reported92.9%
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 MAI-Thinking-1 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

SpecMAI-Thinking-1Phi-4 ReasoningEdge
Context window256K tokens33K tokensMAI-Thinking-1
Max output64K tokens33K tokensMAI-Thinking-1
Input price / 1M
Output price / 1M
Cached input / 1M
Input + output / 1M
Lower is cheaper. List prices; batch, tool and regional fees excluded.
ModalitiestexttextTie
Released2026-08-122025-04-30
Cited benchmark scores154
Reliability

Microsoft status

All providers
Built by Respan
Which one wins on your data?

Public benchmarks are a starting point. Run MAI-Thinking-1 and Phi-4 Reasoning on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.