Model comparison
GPT-3.5-turbo vs Grok-2 (Dec 2024)
Grok-2 (Dec 2024) is the stronger model overall, scoring 33.4 to 23.1 on the Noometry Index.
Last verified . 23 shared benchmarks.
Summary
- They share 23 benchmarks with published results for both. GPT-3.5-turbo scores higher in 0 categories and Grok-2 (Dec 2024) in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Grok-2 (Dec 2024) leads 48.8 to 25.4.
- The biggest single-benchmark swing is MATH Level 5: 15.9% for GPT-3.5-turbo and 63.5% for Grok-2 (Dec 2024).
Side by side
| GPT-3.5-turbo | Grok-2 (Dec 2024) | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 23.1 | 33.4 |
| Released | 2023-03-01 | 2024-08-13 |
| Weights | Proprietary | Proprietary |
| Context window | 16K | — |
| Max output | 4K | — |
| Input $ / M tokens | $0.50 | — |
| Output $ / M tokens | $1.50 | — |
| Results tracked | 44 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok-2 (Dec 2024) leads
GPT-3.5-turbo: 23.5 (#326), Grok-2 (Dec 2024): 32.9 (#254)
| Benchmark | GPT-3.5-turbo | Grok-2 (Dec 2024) |
|---|---|---|
| WeirdML | 3.5% | 22.2% |
| LMArena Coding | 1137 | 1288 |
| BigCodeBench Instruct | 39.1% | — |
| LiveBench Coding | — | 46.4% |
| BigCodeBench Complete | 50.6% | — |
| HumanEval+ | 70.7% | — |
| MBPP+ | 69.7% | — |
Agentic & Tool Use Not comparable
GPT-3.5-turbo: —, Grok-2 (Dec 2024): —
| Benchmark | GPT-3.5-turbo | Grok-2 (Dec 2024) |
|---|---|---|
| METR Time Horizons | 21.5% | — |
Reasoning Grok-2 (Dec 2024) leads
GPT-3.5-turbo: 13.3 (#328), Grok-2 (Dec 2024): 16.4 (#294)
| Benchmark | GPT-3.5-turbo | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Hard Prompts | 1108 | 1272 |
| DTBench | 48.5% | 65.2% |
| Epoch Capabilities Index | 118.48 | 130.48 |
| SimpleBench | — | 22.7% |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | — | 54.8% |
| Mystery Game Puzzles | 3% | — |
| LiveBench Data Analysis | — | 54.5% |
| LMCA | 9.7% | — |
| Adversarial NLI | 58.1% | — |
| BIG-Bench Hard | 61.6% | — |
| CommonsenseQA 2.0 | 57% | — |
| ForecastBench | 50.4 | — |
| LiveBench | — | 54.3% |
| WinoGrande | 81.6% | — |
Math Grok-2 (Dec 2024) leads
GPT-3.5-turbo: 6.0 (#321), Grok-2 (Dec 2024): 19.6 (#278)
| Benchmark | GPT-3.5-turbo | Grok-2 (Dec 2024) |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.2% | 11.5% |
| LMArena Math | 1142 | 1283 |
| MATH Level 5 | 15.9% | 63.5% |
| FrontierMath (Tiers 1-3) | 0% | — |
| LiveBench Math | — | 54.9% |
| FrontierMath (Feb 2025 set) | — | 0.7% |
| GSM8K | 57.8% | — |
Knowledge Grok-2 (Dec 2024) leads
GPT-3.5-turbo: 10.0 (#298), Grok-2 (Dec 2024): 29.8 (#225)
| Benchmark | GPT-3.5-turbo | Grok-2 (Dec 2024) |
|---|---|---|
| GPQA Diamond | 28% | 53.8% |
| LMArena Expert | 1070 | 1254 |
| Confabulations | — | 20.1% |
| ARC (AI2) Challenge | 87.4% | — |
| BoolQ | 87% | — |
| MMLU | 71.4% | — |
| OpenBookQA | 86% | — |
| TriviaQA | 85.8% | — |
Multilingual Grok-2 (Dec 2024) leads
GPT-3.5-turbo: 31.6 (#253), Grok-2 (Dec 2024): 43.3 (#183)
| Benchmark | GPT-3.5-turbo | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Non-English | 1108 | 1282 |
| LMArena Chinese | 1075 | 1289 |
| LMArena French | 1117 | 1317 |
| LMArena German | 1090 | 1287 |
| LMArena Japanese | 1043 | 1244 |
| LMArena Korean | 1019 | 1237 |
| LMArena Russian | 1123 | 1286 |
| LMArena Spanish | 1120 | 1281 |
Instruction Following Grok-2 (Dec 2024) leads
GPT-3.5-turbo: 58.1 (#257), Grok-2 (Dec 2024): 67.1 (#197)
| Benchmark | GPT-3.5-turbo | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Instruction Following | 1119 | 1270 |
| LiveBench Instruction Following | — | 69.6% |
Long Context Grok-2 (Dec 2024) leads
GPT-3.5-turbo: 34.0 (#249), Grok-2 (Dec 2024): 38.7 (#186)
| Benchmark | GPT-3.5-turbo | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Longer Query | 1121 | 1276 |
Writing & Preference Grok-2 (Dec 2024) leads
GPT-3.5-turbo: 25.4 (#300), Grok-2 (Dec 2024): 48.8 (#192)
| Benchmark | GPT-3.5-turbo | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Text | 1125 | 1305 |
| LMArena Creative Writing | 1092 | 1284 |
| LMArena Multi-Turn | 1117 | 1290 |
| Short-Story Creative Writing | — | 63.6% |
| EQ-Bench Creative Writing | 451 | — |
| LiveBench Language | — | 45.6% |
Frequently asked questions
Is GPT-3.5-turbo better than Grok-2 (Dec 2024)?
Grok-2 (Dec 2024) is the stronger model overall, scoring 33.4 to 23.1 on the Noometry Index.
Is GPT-3.5-turbo or Grok-2 (Dec 2024) better for coding?
Grok-2 (Dec 2024) scores higher on coding benchmarks: 32.9 versus 23.5 in the Noometry coding category.
How many benchmarks do GPT-3.5-turbo and Grok-2 (Dec 2024) share?
23 benchmarks have published results for both models. GPT-3.5-turbo has 44 scored results on Noometry and Grok-2 (Dec 2024) has 34.