Model comparison
Grok 4.3 vs Qwen2-72B
Grok 4.3 is the stronger model overall, scoring 43.8 to 30.0 on the Noometry Index.
Last verified . 20 shared benchmarks.
Alibaba (Qwen)Summary
- They share 20 benchmarks with published results for both. Grok 4.3 scores higher in 9 categories and Qwen2-72B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Grok 4.3 leads 52.5 to 21.2.
- The biggest single-benchmark swing is GPQA Diamond: 88.8% for Grok 4.3 and 40.8% for Qwen2-72B.
- Qwen2-72B has downloadable open weights; the other is API-only.
Side by side
| Grok 4.3 | Qwen2-72B | |
|---|---|---|
| Provider | xAI | Alibaba (Qwen) |
| Noometry Index | 43.8 | 30.0 |
| Released | 2026-04-17 | 2024-06-07 |
| Weights | Proprietary | Open |
| Context window | 1M | — |
| Max output | 30K | — |
| Input $ / M tokens | $1.25 | — |
| Output $ / M tokens | $2.50 | — |
| Results tracked | 40 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.3 leads
Grok 4.3: 41.6 (#121), Qwen2-72B: 29.1 (#310)
| Benchmark | Grok 4.3 | Qwen2-72B |
|---|---|---|
| WeirdML | 49.9% | 11.3% |
| LMArena Coding | 1415 | 1196 |
| LMArena WebDev | 1357 | — |
| SciCode | 47.3% | — |
| BigCodeBench Instruct | — | 38.5% |
| BigCodeBench Complete | — | 54% |
| ALE-Bench | 944.17 | — |
Agentic & Tool Use Grok 4.3 leads
Grok 4.3: 27.7 (#99), Qwen2-72B: 17.0 (#146)
| Benchmark | Grok 4.3 | Qwen2-72B |
|---|---|---|
| TheAgentCompany | — | 1.1% |
| GDP.pdf | 8% | — |
| LMArena Search | 1165 | — |
| METR Time Horizons | — | 29.9% |
| Vending-Bench 2 | 35.26 | — |
Reasoning Grok 4.3 leads
Grok 4.3: 35.9 (#68), Qwen2-72B: 23.2 (#181)
| Benchmark | Grok 4.3 | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 1396 | 1191 |
| Epoch Capabilities Index | 149.16 | 125.28 |
| NYT Connections (extended) | 55.2% | — |
| CritPt | 8% | — |
| Chess Puzzles | 25% | — |
| DTBench | 90.7% | — |
| LMCA | 38.3% | — |
| ForecastBench | 60.3 | — |
Math Grok 4.3 leads
Grok 4.3: 46.0 (#74), Qwen2-72B: 30.2 (#236)
| Benchmark | Grok 4.3 | Qwen2-72B |
|---|---|---|
| LMArena Math | 1388 | 1235 |
| FrontierMath (Tiers 1-3) | 42.8% | — |
| FrontierMath Tier 4 | 14.6% | — |
| OTIS Mock AIME 2024-2025 | 93.3% | — |
| ProofBench | 11% | — |
| MATH Level 5 | — | 39.1% |
Knowledge Grok 4.3 leads
Grok 4.3: 52.5 (#62), Qwen2-72B: 21.2 (#275)
| Benchmark | Grok 4.3 | Qwen2-72B |
|---|---|---|
| GPQA Diamond | 88.8% | 40.8% |
| LMArena Expert | 1385 | 1171 |
| SimpleQA Verified | 33.2% | — |
| MMLU | — | 82.4% |
Multimodal Not comparable
Grok 4.3: 31.6 (#104), Qwen2-72B: —
| Benchmark | Grok 4.3 | Qwen2-72B |
|---|---|---|
| LMArena Vision | 1229 | — |
| Blueprint-Bench 2 | 0% | — |
Multilingual Grok 4.3 leads
Grok 4.3: 50.5 (#120), Qwen2-72B: 35.9 (#244)
| Benchmark | Grok 4.3 | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 1385 | 1176 |
| LMArena Chinese | 1422 | 1240 |
| LMArena French | 1412 | 1170 |
| LMArena German | 1395 | 1151 |
| LMArena Japanese | 1379 | 1111 |
| LMArena Korean | 1356 | 1083 |
| LMArena Russian | 1399 | 1169 |
| LMArena Spanish | 1398 | 1169 |
Instruction Following Grok 4.3 leads
Grok 4.3: 72.1 (#140), Qwen2-72B: 61.7 (#241)
| Benchmark | Grok 4.3 | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 1366 | 1181 |
Long Context Grok 4.3 leads
Grok 4.3: 42.5 (#123), Qwen2-72B: 36.1 (#235)
| Benchmark | Grok 4.3 | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | 1393 | 1192 |
Writing & Preference Grok 4.3 leads
Grok 4.3: 58.5 (#118), Qwen2-72B: 40.8 (#241)
| Benchmark | Grok 4.3 | Qwen2-72B |
|---|---|---|
| LMArena Text | 1397 | 1203 |
| LMArena Creative Writing | 1380 | 1181 |
| LMArena Multi-Turn | 1406 | 1196 |
| EQ-Bench 4 | 1075 | — |
Frequently asked questions
Is Grok 4.3 better than Qwen2-72B?
Grok 4.3 is the stronger model overall, scoring 43.8 to 30.0 on the Noometry Index.
Is Grok 4.3 or Qwen2-72B better for coding?
Grok 4.3 scores higher on coding benchmarks: 41.6 versus 29.1 in the Noometry coding category.
How many benchmarks do Grok 4.3 and Qwen2-72B share?
20 benchmarks have published results for both models. Grok 4.3 has 40 scored results on Noometry and Qwen2-72B has 26.