ModelsCompareBest forBenchmarksStatusPricingAPI
Compare / head-to-head

Command AvsPhi-4 Reasoning

Phi-4 Reasoning leads 1 of 1 shared benchmark. Command A has the larger context window (256K tokens).

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

Shared benchmarks
0 – 1
Phi-4 Reasoning leads
Cheaper per token
list price, input + output
Larger context
Command A
256K tokens
Provider uptime (30d)
100% · —
Cohere · Microsoft
Command A
Cohere · released 2025-03-13
text
Context
256K
Max out
8K
Input /1M
$2.5
Output /1M
$10
Cached /1M
Scores
7 · 3 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

BenchmarkCommand APhi-4 ReasoningΔEdge
Reported by both · 1
MMLU-Pro69.6% 74.3% -4.7 ptPhi-4 Reasoning
Only Command A reports · 6
MMLU85.5% not reported
BFCL Overall63.8% not reported
GPQA50.8% not reported
IFEval90.9% not reported
MBPP+86.2% not reported
RepoQA92.6% not reported
Only Phi-4 Reasoning reports · 3
GPQA Diamondnot reported65.8%
AIME 2025not reported62.9%
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 Command A 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

SpecCommand APhi-4 ReasoningEdge
Context window256K tokens33K tokensCommand A
Max output8K tokens33K tokensPhi-4 Reasoning
Input price / 1M$2.5
Output price / 1M$10
Cached input / 1M
Input + output / 1M
Lower is cheaper. List prices; batch, tool and regional fees excluded.
$12.5
ModalitiestexttextTie
Released2025-03-132025-04-30
Cited benchmark scores74
Reliability

Provider status

All providers
More matchups

Command A vs …

Built by Respan
Which one wins on your data?

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