Model comparison
GPT-4 vs Wizardlm 70b
Wizardlm 70b is the stronger model overall, scoring 33.0 to 29.1 on the Noometry Index.
Last verified . 12 shared benchmarks.
MicrosoftSummary
- They share 12 benchmarks with published results for both. GPT-4 scores higher in 5 categories and Wizardlm 70b in 2 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in math, where Wizardlm 70b leads 32.2 to 10.8.
- Wizardlm 70b has downloadable open weights; the other is API-only.
Side by side
| GPT-4 | Wizardlm 70b | |
|---|---|---|
| Provider | OpenAI | Microsoft |
| Noometry Index | 29.1 | 33.0 |
| Released | 2023-03-14 | — |
| Weights | Proprietary | Open |
| Context window | 8K | — |
| Max output | 8K | — |
| Input $ / M tokens | $30 | — |
| Output $ / M tokens | $60 | — |
| Results tracked | 38 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
GPT-4: 31.6 (#283), Wizardlm 70b: 31.4 (#285)
| Benchmark | GPT-4 | Wizardlm 70b |
|---|---|---|
| LMArena Coding | 1254 | 1081 |
| WeirdML | 12.4% | — |
| BigCodeBench Instruct | 46% | — |
| BigCodeBench Complete | 57.2% | — |
| HumanEval+ | 79.3% | — |
Agentic & Tool Use Not comparable
GPT-4: —, Wizardlm 70b: —
| Benchmark | GPT-4 | Wizardlm 70b |
|---|---|---|
| METR Time Horizons | 36.1% | — |
Reasoning Wizardlm 70b leads
GPT-4: 17.8 (#289), Wizardlm 70b: 20.7 (#234)
| Benchmark | GPT-4 | Wizardlm 70b |
|---|---|---|
| LMArena Hard Prompts | 1241 | 1079 |
| Chess Puzzles | 4% | — |
| Mystery Game Puzzles | 12% | — |
| DTBench | 62.7% | — |
| LMCA | 17.1% | — |
| BIG-Bench Hard | 75.1% | — |
| Epoch Capabilities Index | 125.89 | — |
| ForecastBench | 57.8 | — |
| HellaSwag | 95.3% | — |
| WinoGrande | 87.5% | — |
Math Wizardlm 70b leads
GPT-4: 10.8 (#309), Wizardlm 70b: 32.2 (#218)
| Benchmark | GPT-4 | Wizardlm 70b |
|---|---|---|
| LMArena Math | 1269 | 1116 |
| OTIS Mock AIME 2024-2025 | 1.1% | — |
| MATH Level 5 | 23% | — |
| GSM8K | 92% | — |
Knowledge Not comparable
GPT-4: 18.4 (#282), Wizardlm 70b: —
| Benchmark | GPT-4 | Wizardlm 70b |
|---|---|---|
| GPQA Diamond | 35.7% | — |
| LMArena Expert | 1211 | — |
| MMLU | 86.4% | — |
| TriviaQA | 84.8% | — |
Multilingual GPT-4 leads
GPT-4: 40.6 (#215), Wizardlm 70b: 29.6 (#265)
| Benchmark | GPT-4 | Wizardlm 70b |
|---|---|---|
| LMArena Non-English | 1246 | 1078 |
| LMArena Chinese | 1242 | 1052 |
| LMArena German | 1251 | 1083 |
| LMArena Russian | 1251 | 1155 |
| LMArena French | 1283 | — |
| LMArena Japanese | 1209 | — |
| LMArena Korean | 1184 | — |
| LMArena Spanish | 1261 | — |
Instruction Following GPT-4 leads
GPT-4: 65.3 (#222), Wizardlm 70b: 56.3 (#273)
| Benchmark | GPT-4 | Wizardlm 70b |
|---|---|---|
| LMArena Instruction Following | 1241 | 1093 |
Long Context GPT-4 leads
GPT-4: 37.7 (#212), Wizardlm 70b: 33.3 (#263)
| Benchmark | GPT-4 | Wizardlm 70b |
|---|---|---|
| LMArena Longer Query | 1244 | 1097 |
Writing & Preference Too close to call
GPT-4: 34.9 (#268), Wizardlm 70b: 34.8 (#269)
| Benchmark | GPT-4 | Wizardlm 70b |
|---|---|---|
| LMArena Text | 1263 | 1120 |
| LMArena Creative Writing | 1244 | 1149 |
| LMArena Multi-Turn | 1257 | 1108 |
| EQ-Bench Creative Writing | 752 | — |
Frequently asked questions
Is GPT-4 better than Wizardlm 70b?
Wizardlm 70b is the stronger model overall, scoring 33.0 to 29.1 on the Noometry Index.
Is GPT-4 or Wizardlm 70b better for coding?
They score almost the same on coding (31.6 vs 31.4); test both on your own repository before choosing.
How many benchmarks do GPT-4 and Wizardlm 70b share?
12 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Wizardlm 70b has 12.