Model comparison
Granite 3.0 2b Instruct vs phi-3-medium 14B
Granite 3.0 2b Instruct is the stronger model overall, scoring 30.2 to 29.1 on the Noometry Index.
Last verified . 1 shared benchmarks.
MicrosoftSummary
- They share 1 benchmark with published results for both. Granite 3.0 2b Instruct scores higher in 2 categories and phi-3-medium 14B in 1 category; 3 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Granite 3.0 2b Instruct leads 28.4 to 9.1.
- The biggest single-benchmark swing is BigCodeBench Instruct: 20.5% for Granite 3.0 2b Instruct and 37.6% for phi-3-medium 14B.
Side by side
| Granite 3.0 2b Instruct | phi-3-medium 14B | |
|---|---|---|
| Provider | IBM | Microsoft |
| Noometry Index | 30.2 | 29.1 |
| Released | — | 2024-04-23 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 13 | 13 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding phi-3-medium 14B leads
Granite 3.0 2b Instruct: 27.8 (#311), phi-3-medium 14B: 36.3 (#198)
| Benchmark | Granite 3.0 2b Instruct | phi-3-medium 14B |
|---|---|---|
| BigCodeBench Instruct | 20.5% | 37.6% |
| LMArena Coding | 1090 | — |
| BigCodeBench Complete | — | 48.7% |
Reasoning Not comparable
Granite 3.0 2b Instruct: 19.6 (#234), phi-3-medium 14B: —
| Benchmark | Granite 3.0 2b Instruct | phi-3-medium 14B |
|---|---|---|
| LMArena Hard Prompts | 1073 | — |
| Adversarial NLI | — | 55.8% |
| BIG-Bench Hard | — | 81.4% |
| Epoch Capabilities Index | — | 121.18 |
| HellaSwag | — | 82.4% |
| WinoGrande | — | 81.5% |
Math Granite 3.0 2b Instruct leads
Granite 3.0 2b Instruct: 29.7 (#215), phi-3-medium 14B: 25.2 (#249)
| Benchmark | Granite 3.0 2b Instruct | phi-3-medium 14B |
|---|---|---|
| LMArena Math | 1117 | — |
| MATH Level 5 | — | 17.6% |
Knowledge Granite 3.0 2b Instruct leads
Granite 3.0 2b Instruct: 28.4 (#238), phi-3-medium 14B: 9.1 (#301)
| Benchmark | Granite 3.0 2b Instruct | phi-3-medium 14B |
|---|---|---|
| GPQA Diamond | — | 27.6% |
| LMArena Expert | 1064 | — |
| ARC (AI2) Challenge | — | 91.6% |
| MMLU | — | 78% |
| OpenBookQA | — | 87.4% |
| TriviaQA | — | 73.9% |
Multilingual Not comparable
Granite 3.0 2b Instruct: 27.1 (#273), phi-3-medium 14B: —
| Benchmark | Granite 3.0 2b Instruct | phi-3-medium 14B |
|---|---|---|
| LMArena Non-English | 1033 | — |
| LMArena Chinese | 1070 | — |
| LMArena Russian | 1045 | — |
Instruction Following Not comparable
Granite 3.0 2b Instruct: 54.2 (#279), phi-3-medium 14B: —
| Benchmark | Granite 3.0 2b Instruct | phi-3-medium 14B |
|---|---|---|
| LMArena Instruction Following | 1056 | — |
Long Context Not comparable
Granite 3.0 2b Instruct: 32.5 (#263), phi-3-medium 14B: —
| Benchmark | Granite 3.0 2b Instruct | phi-3-medium 14B |
|---|---|---|
| LMArena Longer Query | 1071 | — |
Writing & Preference Not comparable
Granite 3.0 2b Instruct: 29.8 (#287), phi-3-medium 14B: —
| Benchmark | Granite 3.0 2b Instruct | phi-3-medium 14B |
|---|---|---|
| LMArena Text | 1080 | — |
| LMArena Creative Writing | 1046 | — |
| LMArena Multi-Turn | 1053 | — |
Frequently asked questions
Is Granite 3.0 2b Instruct better than phi-3-medium 14B?
Granite 3.0 2b Instruct is the stronger model overall, scoring 30.2 to 29.1 on the Noometry Index.
Is Granite 3.0 2b Instruct or phi-3-medium 14B better for coding?
phi-3-medium 14B scores higher on coding benchmarks: 36.3 versus 27.8 in the Noometry coding category.
How many benchmarks do Granite 3.0 2b Instruct and phi-3-medium 14B share?
1 benchmark has published results for both models. Granite 3.0 2b Instruct has 13 scored results on Noometry and phi-3-medium 14B has 13.