Compare · ModelsLive · 2 picked · head to head
Phi 3 Mini 4k Instruct vs Magnum v4 72B
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Magnum v4 72B wins on 3/6 benchmarks
Magnum v4 72B wins 3 of 6 shared benchmarks. Leads in language · math · reasoning.
Category leads
general·Phi 3 Mini 4k Instructknowledge·Phi 3 Mini 4k Instructlanguage·Magnum v4 72Bmath·Magnum v4 72Breasoning·Magnum v4 72B
Hype vs Reality
Attention vs performance
Phi 3 Mini 4k Instruct
#196 by perf·no signal
Magnum v4 72B
#194 by perf·no signal
Best value
Magnum v4 72B
Phi 3 Mini 4k Instruct
—
no price
Magnum v4 72B
7.0 pts/$
$4.00/M
Vendor risk
Who is behind the model
Microsoft
$3.00T·Big Tech
anthracite-org
private · undisclosed
Head to head
6 benchmarks · 2 models
Phi 3 Mini 4k InstructMagnum v4 72B
BBH (HuggingFace)
Phi 3 Mini 4k Instruct leads by +1.0
Phi 3 Mini 4k Instruct
36.6
Magnum v4 72B
35.5
GPQA
Phi 3 Mini 4k Instruct leads by +0.6
Phi 3 Mini 4k Instruct
11.0
Magnum v4 72B
10.4
IFEval
Magnum v4 72B leads by +1.5
Phi 3 Mini 4k Instruct
54.8
Magnum v4 72B
56.3
MATH Level 5
Magnum v4 72B leads by +3.6
Phi 3 Mini 4k Instruct
16.4
Magnum v4 72B
20.0
MMLU-PRO
Phi 3 Mini 4k Instruct leads by +2.1
Phi 3 Mini 4k Instruct
33.6
Magnum v4 72B
31.4
MUSR
Magnum v4 72B leads by +0.3
Phi 3 Mini 4k Instruct
13.1
Magnum v4 72B
13.4
Full benchmark table
| Benchmark | Phi 3 Mini 4k Instruct | Magnum v4 72B |
|---|---|---|
BBH (HuggingFace) | 36.6 | 35.5 |
GPQA | 11.0 | 10.4 |
IFEval | 54.8 | 56.3 |
MATH Level 5 | 16.4 | 20.0 |
MMLU-PRO | 33.6 | 31.4 |
MUSR | 13.1 | 13.4 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| Phi 3 Mini 4k Instruct | — | — | — | — |
| Magnum v4 72B | $3.00 | $5.00 | 16K tokens (~8 books) | $35.00 |