Model comparison
Mercury 2 vs Phi-4 Mini
Mercury 2 is the stronger model overall, scoring 39.1 to 30.9 on the Noometry Index. Phi-4 Mini costs 2.9× less per token, which makes it the better buy when Mercury 2's lead doesn't matter for your workload.
Last verified . 3 shared benchmarks.
MicrosoftSummary
- They share 3 benchmarks with published results for both. Mercury 2 scores higher in 3 categories and Phi-4 Mini in 0 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Mercury 2 leads 36.2 to 25.3.
- The biggest single-benchmark swing is SciCode: 38.7% for Mercury 2 and 10.8% for Phi-4 Mini.
- Phi-4 Mini is cheaper at $0.075 / $0.30 per million input/output tokens, against $0.25 / $0.75 for Mercury 2.
- Phi-4 Mini has downloadable open weights; the other is API-only.
Side by side
| Mercury 2 | Phi-4 Mini | |
|---|---|---|
| Provider | Inception | Microsoft |
| Noometry Index | 39.1 | 30.9 |
| Released | 2026-02-20 | 2024-12-11 |
| Weights | Proprietary | Open |
| Context window | 128K | 128K |
| Max output | 50K | 4K |
| Input $ / M tokens | $0.25 | $0.075 |
| Output $ / M tokens | $0.75 | $0.30 |
| Results tracked | 17 | 3 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mercury 2 leads
Mercury 2: 33.5 (#255), Phi-4 Mini: 28.1 (#317)
| Benchmark | Mercury 2 | Phi-4 Mini |
|---|---|---|
| SciCode | 38.7% | 10.8% |
| LMArena WebDev | 1171 | — |
| WeirdML | 43.2% | — |
| LMArena Coding | 1391 | — |
| ALE-Bench | 785.58 | — |
Reasoning Mercury 2 leads
Mercury 2: 23.8 (#170), Phi-4 Mini: 22.4 (#195)
| Benchmark | Mercury 2 | Phi-4 Mini |
|---|---|---|
| CritPt | 0.8% | 0% |
| LMArena Hard Prompts | 1362 | — |
Knowledge Mercury 2 leads
Mercury 2: 36.2 (#172), Phi-4 Mini: 25.3 (#262)
| Benchmark | Mercury 2 | Phi-4 Mini |
|---|---|---|
| Vectara Hallucination Rate | 12.3% | 23.5% |
| LMArena Expert | 1358 | — |
Multilingual Not comparable
Mercury 2: 46.6 (#157), Phi-4 Mini: —
| Benchmark | Mercury 2 | Phi-4 Mini |
|---|---|---|
| LMArena Non-English | 1331 | — |
| LMArena Chinese | 1417 | — |
| LMArena Russian | 1304 | — |
Instruction Following Not comparable
Mercury 2: 70.2 (#165), Phi-4 Mini: —
| Benchmark | Mercury 2 | Phi-4 Mini |
|---|---|---|
| LMArena Instruction Following | 1329 | — |
Long Context Not comparable
Mercury 2: 40.5 (#154), Phi-4 Mini: —
| Benchmark | Mercury 2 | Phi-4 Mini |
|---|---|---|
| LMArena Longer Query | 1330 | — |
Writing & Preference Not comparable
Mercury 2: 53.8 (#155), Phi-4 Mini: —
| Benchmark | Mercury 2 | Phi-4 Mini |
|---|---|---|
| LMArena Text | 1355 | — |
| LMArena Creative Writing | 1289 | — |
| LMArena Multi-Turn | 1358 | — |
Frequently asked questions
Is Mercury 2 better than Phi-4 Mini?
Mercury 2 is the stronger model overall, scoring 39.1 to 30.9 on the Noometry Index. Phi-4 Mini costs 2.9× less per token, which makes it the better buy when Mercury 2's lead doesn't matter for your workload.
Which is cheaper, Mercury 2 or Phi-4 Mini?
Phi-4 Mini is cheaper. It lists at $0.075 per million input tokens and $0.30 per million output tokens; Mercury 2 lists at $0.25 and $0.75.
Is Mercury 2 or Phi-4 Mini better for coding?
Mercury 2 scores higher on coding benchmarks: 33.5 versus 28.1 in the Noometry coding category.
Which has the bigger context window?
Both accept 128K tokens.
How many benchmarks do Mercury 2 and Phi-4 Mini share?
3 benchmarks have published results for both models. Mercury 2 has 17 scored results on Noometry and Phi-4 Mini has 3.