Model comparison
Grok 3 vs Hy4 preview
Hy4 preview is the stronger model overall, scoring 44.6 to 39.6 on the Noometry Index.
Last verified . 0 shared benchmarks.
TencentSummary
- The widest gap is in math, where Hy4 preview leads 54.5 to 36.2.
- Hy4 preview has downloadable open weights; the other is API-only.
Side by side
| Grok 3 | Hy4 preview | |
|---|---|---|
| Provider | xAI | Tencent |
| Noometry Index | 39.6 | 44.6 |
| Released | 2025-04-09 | 2026-08-28 |
| Weights | Proprietary | Open |
| Context window | — | 1.05M |
| Max output | — | 64K |
| Input $ / M tokens | — | $0.83 |
| Output $ / M tokens | — | $2.50 |
| Results tracked | 40 | 3 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Hy4 preview leads
Grok 3: 41.5 (#113), Hy4 preview: 51.4 (#36)
| Benchmark | Grok 3 | Hy4 preview |
|---|---|---|
| Aider Polyglot | 53.3% | — |
| LMArena WebDev | — | 1633 |
| WeirdML | 37.2% | — |
| LMArena Coding | 1432 | — |
Agentic & Tool Use Not comparable
Grok 3: 30.2 (#69), Hy4 preview: —
| Benchmark | Grok 3 | Hy4 preview |
|---|---|---|
| BALROG | 29.5% | — |
Reasoning Hy4 preview leads
Grok 3: 13.8 (#322), Hy4 preview: 30.6 (#75)
| Benchmark | Grok 3 | Hy4 preview |
|---|---|---|
| ARC-AGI-2 | 0% | — |
| SimpleBench | 36.1% | — |
| Kagi LLM Benchmark | 61.3% | — |
| NYT Connections (extended) | — | 68.2% |
| ARC-AGI-1 | 5.5% | — |
| LMArena Hard Prompts | 1433 | — |
| Epoch Capabilities Index | 138.34 | — |
Math Hy4 preview leads
Grok 3: 36.2 (#122), Hy4 preview: 54.5 (#37)
| Benchmark | Grok 3 | Hy4 preview |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 55.6% | — |
| ProofBench | — | 75% |
| Omni-MATH | 46.4% | — |
| LMArena Math | 1390 | — |
| MATH Level 5 | 88.7% | — |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Not comparable
Grok 3: 46.0 (#76), Hy4 preview: —
| Benchmark | Grok 3 | Hy4 preview |
|---|---|---|
| GPQA Diamond | 75.8% | — |
| MMLU-Pro | 78.8% | — |
| Confabulations | 14.2% | — |
| Vectara Hallucination Rate | 5.8% | — |
| GPQA (HELM) | 65% | — |
| LMArena Expert | 1420 | — |
Multilingual Not comparable
Grok 3: 52.5 (#81), Hy4 preview: —
| Benchmark | Grok 3 | Hy4 preview |
|---|---|---|
| LMArena Non-English | 1411 | — |
| LMArena Chinese | 1448 | — |
| LMArena French | 1460 | — |
| LMArena German | 1431 | — |
| LMArena Japanese | 1389 | — |
| LMArena Korean | 1374 | — |
| LMArena Russian | 1416 | — |
| LMArena Spanish | 1417 | — |
Instruction Following Not comparable
Grok 3: 75.2 (#69), Hy4 preview: —
| Benchmark | Grok 3 | Hy4 preview |
|---|---|---|
| IFEval | 88.4% | — |
| LMArena Instruction Following | 1409 | — |
Long Context Not comparable
Grok 3: 38.7 (#187), Hy4 preview: —
| Benchmark | Grok 3 | Hy4 preview |
|---|---|---|
| Fiction.LiveBench | 58.3% | — |
| LMArena Longer Query | 1439 | — |
Writing & Preference Not comparable
Grok 3: 56.0 (#136), Hy4 preview: —
| Benchmark | Grok 3 | Hy4 preview |
|---|---|---|
| LMArena Text | 1426 | — |
| LMArena Creative Writing | 1414 | — |
| Short-Story Creative Writing | 76.4% | — |
| EQ-Bench Creative Writing | 1186 | — |
| WildBench | 84.9% | — |
| LMArena Multi-Turn | 1424 | — |
Frequently asked questions
Is Grok 3 better than Hy4 preview?
Hy4 preview is the stronger model overall, scoring 44.6 to 39.6 on the Noometry Index.
Is Grok 3 or Hy4 preview better for coding?
Hy4 preview scores higher on coding benchmarks: 51.4 versus 41.5 in the Noometry coding category.
How many benchmarks do Grok 3 and Hy4 preview share?
0 benchmarks have published results for both models. Grok 3 has 40 scored results on Noometry and Hy4 preview has 3.