Compare · ModelsLive · 2 picked · head to head
Magnum v4 72B vs Phi 3.5 Mini Instruct
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Phi 3.5 Mini Instruct wins on 4/6 benchmarks
Phi 3.5 Mini Instruct wins 4 of 6 shared benchmarks. Leads in general · knowledge · language.
Category leads
general·Phi 3.5 Mini Instructknowledge·Phi 3.5 Mini Instructlanguage·Phi 3.5 Mini Instructmath·Magnum v4 72Breasoning·Magnum v4 72B
Hype vs Reality
Attention vs performance
Magnum v4 72B
#277 by perf·no signal
Phi 3.5 Mini Instruct
#274 by perf·no signal
Best value
Magnum v4 72B
Magnum v4 72B
7.4 pts/$
$3.75/M
Phi 3.5 Mini Instruct
n/a
no price
Vendor risk
Who is behind the model
anthracite-org
private · undisclosed
Microsoft
$3.96T·Big Tech
Head to head
6 benchmarks · 2 models
Magnum v4 72BPhi 3.5 Mini Instruct
BBH (HuggingFace)
Phi 3.5 Mini Instruct leads by +1.2
Magnum v4 72B
35.5
Phi 3.5 Mini Instruct
36.8
GPQA
Phi 3.5 Mini Instruct leads by +1.6
Magnum v4 72B
10.4
Phi 3.5 Mini Instruct
12.0
IFEval
Phi 3.5 Mini Instruct leads by +1.5
Magnum v4 72B
56.3
Phi 3.5 Mini Instruct
57.8
MATH Level 5
Magnum v4 72B leads by +0.4
Magnum v4 72B
20.0
Phi 3.5 Mini Instruct
19.6
MMLU-PRO
Phi 3.5 Mini Instruct leads by +1.5
Magnum v4 72B
31.4
Phi 3.5 Mini Instruct
32.9
MUSR
Magnum v4 72B leads by +3.3
Magnum v4 72B
13.4
Phi 3.5 Mini Instruct
10.1
Full benchmark table
| Benchmark | Magnum v4 72B | Phi 3.5 Mini Instruct |
|---|---|---|
BBH (HuggingFace) | 35.5 | 36.8 |
GPQA | 10.4 | 12.0 |
IFEval | 56.3 | 57.8 |
MATH Level 5 | 20.0 | 19.6 |
MMLU-PRO | 31.4 | 32.9 |
MUSR | 13.4 | 10.1 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $2.50 | $5.00 | 33K tokens (~16 books) | $31.25 | |
| n/a | n/a | n/a | n/a |