Model comparison
GPT-4 Turbo vs Phi 3 Medium 4k Instruct
Phi 3 Medium 4k Instruct is the stronger model overall, scoring 33.0 to 30.5 on the Noometry Index.
Last verified . 17 shared benchmarks.
MicrosoftSummary
- They share 17 benchmarks with published results for both. GPT-4 Turbo scores higher in 5 categories and Phi 3 Medium 4k Instruct in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Phi 3 Medium 4k Instruct leads 33.4 to 9.0.
- Phi 3 Medium 4k Instruct has downloadable open weights; the other is API-only.
Side by side
| GPT-4 Turbo | Phi 3 Medium 4k Instruct | |
|---|---|---|
| Provider | OpenAI | Microsoft |
| Noometry Index | 30.5 | 33.0 |
| Released | 2023-11-06 | — |
| Weights | Proprietary | Open |
| Context window | 128K | — |
| Max output | 4K | — |
| Input $ / M tokens | $10 | — |
| Output $ / M tokens | $30 | — |
| Results tracked | 36 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
GPT-4 Turbo: 33.8 (#249), Phi 3 Medium 4k Instruct: 32.9 (#268)
| Benchmark | GPT-4 Turbo | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Coding | 1268 | 1130 |
| WeirdML | 18% | — |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 58.2% | — |
| HumanEval+ | 86.6% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use Not comparable
GPT-4 Turbo: —, Phi 3 Medium 4k Instruct: —
| Benchmark | GPT-4 Turbo | Phi 3 Medium 4k Instruct |
|---|---|---|
| METR Time Horizons | 36.7% | — |
Reasoning Phi 3 Medium 4k Instruct leads
GPT-4 Turbo: 15.3 (#317), Phi 3 Medium 4k Instruct: 21.7 (#214)
| Benchmark | GPT-4 Turbo | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Hard Prompts | 1251 | 1127 |
| SimpleBench | 25.1% | — |
| Chess Puzzles | 6% | — |
| DTBench | 61.6% | — |
| LMCA | 9.8% | — |
| Epoch Capabilities Index | 127.25 | — |
| ForecastBench | 59.4 | — |
Math Phi 3 Medium 4k Instruct leads
GPT-4 Turbo: 9.0 (#322), Phi 3 Medium 4k Instruct: 33.4 (#202)
| Benchmark | GPT-4 Turbo | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Math | 1272 | 1173 |
| FrontierMath (Tiers 1-3) | 0.7% | — |
| OTIS Mock AIME 2024-2025 | 6.7% | — |
| MATH Level 5 | 46.7% | — |
Knowledge Phi 3 Medium 4k Instruct leads
GPT-4 Turbo: 24.3 (#268), Phi 3 Medium 4k Instruct: 30.2 (#229)
| Benchmark | GPT-4 Turbo | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Expert | 1223 | 1108 |
| GPQA Diamond | 46.6% | — |
| Confabulations | 28.4% | — |
| MMLU | 81.3% | — |
Multimodal Not comparable
GPT-4 Turbo: 30.6 (#110), Phi 3 Medium 4k Instruct: —
| Benchmark | GPT-4 Turbo | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Vision | 1090 | — |
Multilingual GPT-4 Turbo leads
GPT-4 Turbo: 40.5 (#216), Phi 3 Medium 4k Instruct: 30.6 (#263)
| Benchmark | GPT-4 Turbo | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Non-English | 1245 | 1094 |
| LMArena Chinese | 1242 | 1108 |
| LMArena French | 1276 | 1110 |
| LMArena German | 1259 | 1101 |
| LMArena Japanese | 1194 | 1042 |
| LMArena Korean | 1187 | 954 |
| LMArena Russian | 1259 | 1145 |
| LMArena Spanish | 1260 | 1095 |
Instruction Following GPT-4 Turbo leads
GPT-4 Turbo: 65.8 (#216), Phi 3 Medium 4k Instruct: 57.6 (#268)
| Benchmark | GPT-4 Turbo | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Instruction Following | 1249 | 1113 |
Long Context GPT-4 Turbo leads
GPT-4 Turbo: 38.0 (#206), Phi 3 Medium 4k Instruct: 34.0 (#255)
| Benchmark | GPT-4 Turbo | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1254 | 1120 |
Writing & Preference GPT-4 Turbo leads
GPT-4 Turbo: 47.7 (#206), Phi 3 Medium 4k Instruct: 34.1 (#272)
| Benchmark | GPT-4 Turbo | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Text | 1272 | 1138 |
| LMArena Creative Writing | 1269 | 1107 |
| LMArena Multi-Turn | 1267 | 1088 |
Frequently asked questions
Is GPT-4 Turbo better than Phi 3 Medium 4k Instruct?
Phi 3 Medium 4k Instruct is the stronger model overall, scoring 33.0 to 30.5 on the Noometry Index.
Is GPT-4 Turbo or Phi 3 Medium 4k Instruct better for coding?
They score almost the same on coding (33.8 vs 32.9); test both on your own repository before choosing.
How many benchmarks do GPT-4 Turbo and Phi 3 Medium 4k Instruct share?
17 benchmarks have published results for both models. GPT-4 Turbo has 36 scored results on Noometry and Phi 3 Medium 4k Instruct has 17.