Model comparison
Claude Sonnet 4 vs Hy4 preview
Hy4 preview is the stronger model overall, scoring 44.6 to 40.6 on the Noometry Index.
Last verified . 0 shared benchmarks.
TencentSummary
- The widest gap is in math, where Hy4 preview leads 54.5 to 41.5.
- Hy4 preview is cheaper at $0.83 / $2.50 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.
- Hy4 preview accepts more context: 1.05M tokens versus 200K.
- Hy4 preview has downloadable open weights; the other is API-only.
Side by side
| Claude Sonnet 4 | Hy4 preview | |
|---|---|---|
| Provider | Anthropic | Tencent |
| Noometry Index | 40.6 | 44.6 |
| Released | 2025-05-22 | 2026-08-28 |
| Weights | Proprietary | Open |
| Context window | 200K | 1.05M |
| Max output | 64K | 64K |
| Input $ / M tokens | $3 | $0.83 |
| Output $ / M tokens | $15 | $2.50 |
| Results tracked | 58 | 3 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Hy4 preview leads
Claude Sonnet 4: 44.0 (#70), Hy4 preview: 51.4 (#36)
| Benchmark | Claude Sonnet 4 | Hy4 preview |
|---|---|---|
| SWE-bench Verified (bash only) | 64.9% | — |
| Aider Polyglot | 61.3% | — |
| LMArena WebDev | — | 1633 |
| SciCode | 40% | — |
| GSO | 4.9% | — |
| WeirdML | 46.1% | — |
| LMArena Coding | 1414 | — |
| ALE-Bench | 655.35 | — |
Agentic & Tool Use Not comparable
Claude Sonnet 4: 38.3 (#29), Hy4 preview: —
| Benchmark | Claude Sonnet 4 | Hy4 preview |
|---|---|---|
| TheAgentCompany | 33.1% | — |
| Cybench | 35% | — |
| DeepResearch Bench | 46.6% | — |
| OSWorld | 43.9% | — |
| METR Time Horizons | 62% | — |
Reasoning Hy4 preview leads
Claude Sonnet 4: 23.0 (#169), Hy4 preview: 30.6 (#75)
| Benchmark | Claude Sonnet 4 | Hy4 preview |
|---|---|---|
| ARC-AGI-2 | 5.9% | — |
| SimpleBench | 45.5% | — |
| Kagi LLM Benchmark | 73% | — |
| NYT Connections (extended) | — | 68.2% |
| ARC-AGI-1 | 40% | — |
| CritPt | 0.3% | — |
| EnigmaEval | 3.1% | — |
| LMArena Hard Prompts | 1372 | — |
| DTBench | 77.1% | — |
| LMCA | 29% | — |
| Epoch Capabilities Index | 141.69 | — |
| ForecastBench | 60.2 | — |
Math Hy4 preview leads
Claude Sonnet 4: 41.5 (#71), Hy4 preview: 54.5 (#37)
| Benchmark | Claude Sonnet 4 | Hy4 preview |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | — |
| ProofBench | — | 75% |
| Omni-MATH | 60.2% | — |
| LMArena Math | 1375 | — |
| MATH Level 5 | 84.4% | — |
| FrontierMath (Feb 2025 set) | 4.1% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Not comparable
Claude Sonnet 4: 41.6 (#99), Hy4 preview: —
| Benchmark | Claude Sonnet 4 | Hy4 preview |
|---|---|---|
| GPQA Diamond | 79.2% | — |
| Humanity's Last Exam | 7.8% | — |
| MMLU-Pro | 84.3% | — |
| Confabulations | 13.2% | — |
| Vectara Hallucination Rate | 10.3% | — |
| GPQA (HELM) | 70.6% | — |
| LMArena Expert | 1371 | — |
Multimodal Not comparable
Claude Sonnet 4: 25.6 (#116), Hy4 preview: —
| Benchmark | Claude Sonnet 4 | Hy4 preview |
|---|---|---|
| LMArena Vision | 1191 | — |
| GeoBench | 37% | — |
| VPCT | 34% | — |
| MindCube | 44.8% | — |
Multilingual Not comparable
Claude Sonnet 4: 46.9 (#151), Hy4 preview: —
| Benchmark | Claude Sonnet 4 | Hy4 preview |
|---|---|---|
| LMArena Non-English | 1333 | — |
| LMArena Chinese | 1351 | — |
| LMArena French | 1364 | — |
| LMArena German | 1331 | — |
| LMArena Japanese | 1301 | — |
| LMArena Korean | 1293 | — |
| LMArena Russian | 1355 | — |
| LMArena Spanish | 1356 | — |
Instruction Following Not comparable
Claude Sonnet 4: 71.8 (#141), Hy4 preview: —
| Benchmark | Claude Sonnet 4 | Hy4 preview |
|---|---|---|
| IFEval | 84% | — |
| LMArena Instruction Following | 1375 | — |
Long Context Not comparable
Claude Sonnet 4: 33.6 (#254), Hy4 preview: —
| Benchmark | Claude Sonnet 4 | Hy4 preview |
|---|---|---|
| Fiction.LiveBench | 46.9% | — |
| LMArena Longer Query | 1398 | — |
Writing & Preference Not comparable
Claude Sonnet 4: 57.3 (#127), Hy4 preview: —
| Benchmark | Claude Sonnet 4 | Hy4 preview |
|---|---|---|
| LMArena Text | 1351 | — |
| LMArena Creative Writing | 1345 | — |
| Short-Story Creative Writing | 81.4% | — |
| EQ-Bench Creative Writing | 1483 | — |
| WildBench | 83.8% | — |
| LMArena Multi-Turn | 1376 | — |
Frequently asked questions
Is Claude Sonnet 4 better than Hy4 preview?
Hy4 preview is the stronger model overall, scoring 44.6 to 40.6 on the Noometry Index.
Which is cheaper, Claude Sonnet 4 or Hy4 preview?
Hy4 preview is cheaper. It lists at $0.83 per million input tokens and $2.50 per million output tokens; Claude Sonnet 4 lists at $3 and $15.
Is Claude Sonnet 4 or Hy4 preview better for coding?
Hy4 preview scores higher on coding benchmarks: 51.4 versus 44.0 in the Noometry coding category.
Which has the bigger context window?
Hy4 preview does, with 1.05M tokens against 200K.
How many benchmarks do Claude Sonnet 4 and Hy4 preview share?
0 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and Hy4 preview has 3.