Model comparison
Phi-4 vs Yi-34B
Phi-4 is the stronger model overall, scoring 31.2 to 27.8 on the Noometry Index.
Last verified . 21 shared benchmarks.
MicrosoftSummary
- They share 21 benchmarks with published results for both. Phi-4 scores higher in 6 categories and Yi-34B in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Phi-4 leads 32.6 to 7.5.
- The biggest single-benchmark swing is MATH Level 5: 64.9% for Phi-4 and 5.1% for Yi-34B.
Side by side
| Phi-4 | Yi-34B | |
|---|---|---|
| Provider | Microsoft | 01.AI |
| Noometry Index | 31.2 | 27.8 |
| Released | 2024-12-11 | 2023-11-02 |
| Weights | Open | Open |
| Context window | 128K | — |
| Max output | 4K | — |
| Input $ / M tokens | $0.07 | — |
| Output $ / M tokens | $0.14 | — |
| Results tracked | 37 | 23 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Phi-4 leads
Phi-4: 34.4 (#239), Yi-34B: 32.3 (#274)
| Benchmark | Phi-4 | Yi-34B |
|---|---|---|
| LMArena Coding | 1231 | 1112 |
| BigCodeBench Instruct | 45.5% | — |
| LiveBench Coding | 30.7% | — |
| BigCodeBench Complete | 55.4% | — |
Agentic & Tool Use Not comparable
Phi-4: 22.8 (#128), Yi-34B: —
| Benchmark | Phi-4 | Yi-34B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 28.8% | — |
| BALROG | 11.6% | — |
Reasoning Yi-34B leads
Phi-4: 17.7 (#291), Yi-34B: 21.2 (#226)
| Benchmark | Phi-4 | Yi-34B |
|---|---|---|
| LMArena Hard Prompts | 1220 | 1104 |
| Epoch Capabilities Index | 130.42 | 117.39 |
| Chess Puzzles | 1% | — |
| LiveBench Reasoning | 47.8% | — |
| LiveBench Data Analysis | 45.2% | — |
| BIG-Bench Hard | — | 71.7% |
| LiveBench | 41.6% | — |
Math Too close to call
Phi-4: 20.8 (#285), Yi-34B: 21.6 (#282)
| Benchmark | Phi-4 | Yi-34B |
|---|---|---|
| LMArena Math | 1246 | 1114 |
| MATH Level 5 | 64.9% | 5.1% |
| OTIS Mock AIME 2024-2025 | 13.8% | — |
| LiveBench Math | 42% | — |
| GSM8K | — | 76% |
Knowledge Phi-4 leads
Phi-4: 32.6 (#209), Yi-34B: 7.5 (#309)
| Benchmark | Phi-4 | Yi-34B |
|---|---|---|
| GPQA Diamond | 56.1% | 14.7% |
| LMArena Expert | 1203 | 1061 |
| MMLU | 84.8% | 76.3% |
| Confabulations | 29.4% | — |
| Vectara Hallucination Rate | 3.7% | — |
Multilingual Phi-4 leads
Phi-4: 37.2 (#237), Yi-34B: 29.7 (#264)
| Benchmark | Phi-4 | Yi-34B |
|---|---|---|
| LMArena Non-English | 1197 | 1079 |
| LMArena Chinese | 1212 | 1176 |
| LMArena French | 1224 | 1081 |
| LMArena German | 1222 | 1042 |
| LMArena Japanese | 1158 | 993 |
| LMArena Korean | 1151 | 959 |
| LMArena Russian | 1209 | 1050 |
| LMArena Spanish | 1234 | 1070 |
Instruction Following Phi-4 leads
Phi-4: 60.4 (#251), Yi-34B: 56.2 (#274)
| Benchmark | Phi-4 | Yi-34B |
|---|---|---|
| LMArena Instruction Following | 1201 | 1091 |
| LiveBench Instruction Following | 58.4% | — |
Long Context Phi-4 leads
Phi-4: 36.9 (#226), Yi-34B: 33.2 (#264)
| Benchmark | Phi-4 | Yi-34B |
|---|---|---|
| LMArena Longer Query | 1217 | 1094 |
Writing & Preference Phi-4 leads
Phi-4: 40.5 (#244), Yi-34B: 34.1 (#273)
| Benchmark | Phi-4 | Yi-34B |
|---|---|---|
| LMArena Text | 1217 | 1129 |
| LMArena Creative Writing | 1182 | 1108 |
| LMArena Multi-Turn | 1206 | 1113 |
| Short-Story Creative Writing | 62.6% | — |
| LiveBench Language | 25.6% | — |
Frequently asked questions
Is Phi-4 better than Yi-34B?
Phi-4 is the stronger model overall, scoring 31.2 to 27.8 on the Noometry Index.
Is Phi-4 or Yi-34B better for coding?
Phi-4 scores higher on coding benchmarks: 34.4 versus 32.3 in the Noometry coding category.
How many benchmarks do Phi-4 and Yi-34B share?
21 benchmarks have published results for both models. Phi-4 has 37 scored results on Noometry and Yi-34B has 23.