Model comparison
Falcon-180B vs Phi 3 Mini 4k Instruct
Falcon-180B is the stronger model overall, scoring 32.2 to 27.9 on the Noometry Index.
Last verified . 11 shared benchmarks.
Technology Innovation Institute
MicrosoftSummary
- They share 11 benchmarks with published results for both. Falcon-180B scores higher in 3 categories and Phi 3 Mini 4k Instruct in 1 category; 4 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where Falcon-180B leads 53.4 to 47.7.
Side by side
| Falcon-180B | Phi 3 Mini 4k Instruct | |
|---|---|---|
| Provider | Technology Innovation Institute | Microsoft |
| Noometry Index | 32.2 | 27.9 |
| Released | 2023-09-06 | 2024-04-23 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 16 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Falcon-180B: —, Phi 3 Mini 4k Instruct: 26.6 (#323)
| Benchmark | Falcon-180B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LiveBench Coding | — | 15.5% |
| LMArena Coding | — | 1093 |
| HumanEval+ | — | 59.1% |
| MBPP+ | — | 54.2% |
Reasoning Falcon-180B leads
Falcon-180B: 19.1 (#269), Phi 3 Mini 4k Instruct: 14.1 (#328)
| Benchmark | Falcon-180B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Hard Prompts | 1007 | 1072 |
| HellaSwag | 89% | 76.7% |
| WinoGrande | 87.1% | 70.8% |
| Chess Puzzles | — | 0% |
| LiveBench Reasoning | — | 26.8% |
| LiveBench Data Analysis | — | 34.7% |
| Adversarial NLI | — | 52.8% |
| BIG-Bench Hard | — | 71.7% |
| Epoch Capabilities Index | 112.13 | — |
| LAMBADA | 79.8% | — |
| LiveBench | — | 22.4% |
| PIQA | 84.9% | — |
Math Not comparable
Falcon-180B: —, Phi 3 Mini 4k Instruct: 26.6 (#257)
| Benchmark | Falcon-180B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LiveBench Math | — | 15.7% |
| LMArena Math | — | 1111 |
| GSM8K | 54.4% | — |
Knowledge Not comparable
Falcon-180B: —, Phi 3 Mini 4k Instruct: 28.5 (#246)
| Benchmark | Falcon-180B | Phi 3 Mini 4k Instruct |
|---|---|---|
| ARC (AI2) Challenge | 67.8% | 84.9% |
| MMLU | 70.6% | 68.8% |
| OpenBookQA | 64.2% | 88% |
| LMArena Expert | — | 1045 |
| BoolQ | 89% | — |
| TriviaQA | — | 64% |
Multilingual Phi 3 Mini 4k Instruct leads
Falcon-180B: 25.2 (#286), Phi 3 Mini 4k Instruct: 26.3 (#280)
| Benchmark | Falcon-180B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Non-English | 1000 | 1021 |
| LMArena Chinese | — | 1021 |
| LMArena French | — | 1076 |
| LMArena German | — | 1044 |
| LMArena Japanese | — | 935 |
| LMArena Korean | — | 905 |
| LMArena Russian | — | 1022 |
| LMArena Spanish | — | 1085 |
Instruction Following Falcon-180B leads
Falcon-180B: 53.4 (#286), Phi 3 Mini 4k Instruct: 47.7 (#303)
| Benchmark | Falcon-180B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Instruction Following | 1047 | 1053 |
| LiveBench Instruction Following | — | 39.1% |
Long Context Not comparable
Falcon-180B: —, Phi 3 Mini 4k Instruct: 31.7 (#276)
| Benchmark | Falcon-180B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Longer Query | — | 1044 |
Writing & Preference Falcon-180B leads
Falcon-180B: 29.1 (#295), Phi 3 Mini 4k Instruct: 27.6 (#300)
| Benchmark | Falcon-180B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Text | 1054 | 1073 |
| LMArena Creative Writing | 1089 | 1037 |
| LMArena Multi-Turn | 1013 | 1018 |
| LiveBench Language | — | 9.2% |
Frequently asked questions
Is Falcon-180B better than Phi 3 Mini 4k Instruct?
Falcon-180B is the stronger model overall, scoring 32.2 to 27.9 on the Noometry Index.
How many benchmarks do Falcon-180B and Phi 3 Mini 4k Instruct share?
11 benchmarks have published results for both models. Falcon-180B has 16 scored results on Noometry and Phi 3 Mini 4k Instruct has 35.