Model comparison
GPT-4 vs Hy3
Hy3 is the stronger model overall, scoring 44.2 to 29.1 on the Noometry Index.
Last verified . 17 shared benchmarks.
TencentSummary
- They share 17 benchmarks with published results for both. GPT-4 scores higher in 0 categories and Hy3 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Hy3 leads 40.1 to 10.8.
- Hy3 is cheaper at $0.0825 / $0.33 per million input/output tokens, against $30 / $60 for GPT-4.
- Hy3 accepts more context: 262K tokens versus 8K.
- Hy3 has downloadable open weights; the other is API-only.
Side by side
| GPT-4 | Hy3 | |
|---|---|---|
| Provider | OpenAI | Tencent |
| Noometry Index | 29.1 | 44.2 |
| Released | 2023-03-14 | 2026-07-06 |
| Weights | Proprietary | Open |
| Context window | 8K | 262K |
| Max output | 8K | 128K |
| Input $ / M tokens | $30 | $0.0825 |
| Output $ / M tokens | $60 | $0.33 |
| Results tracked | 38 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Hy3 leads
GPT-4: 31.6 (#283), Hy3: 46.8 (#63)
| Benchmark | GPT-4 | Hy3 |
|---|---|---|
| LMArena Coding | 1254 | 1464 |
| LMArena WebDev | — | 1508 |
| WeirdML | 12.4% | — |
| BigCodeBench Instruct | 46% | — |
| BigCodeBench Complete | 57.2% | — |
| HumanEval+ | 79.3% | — |
Agentic & Tool Use Not comparable
GPT-4: —, Hy3: —
| Benchmark | GPT-4 | Hy3 |
|---|---|---|
| METR Time Horizons | 36.1% | — |
Reasoning Hy3 leads
GPT-4: 17.8 (#289), Hy3: 26.1 (#136)
| Benchmark | GPT-4 | Hy3 |
|---|---|---|
| LMArena Hard Prompts | 1241 | 1447 |
| NYT Connections (extended) | — | 41.2% |
| 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 Hy3 leads
GPT-4: 10.8 (#309), Hy3: 40.1 (#93)
| Benchmark | GPT-4 | Hy3 |
|---|---|---|
| LMArena Math | 1269 | 1475 |
| OTIS Mock AIME 2024-2025 | 1.1% | — |
| MATH Level 5 | 23% | — |
| GSM8K | 92% | — |
Knowledge Hy3 leads
GPT-4: 18.4 (#282), Hy3: 40.8 (#114)
| Benchmark | GPT-4 | Hy3 |
|---|---|---|
| LMArena Expert | 1211 | 1460 |
| GPQA Diamond | 35.7% | — |
| MMLU | 86.4% | — |
| TriviaQA | 84.8% | — |
Multilingual Hy3 leads
GPT-4: 40.6 (#215), Hy3: 53.5 (#65)
| Benchmark | GPT-4 | Hy3 |
|---|---|---|
| LMArena Non-English | 1246 | 1426 |
| LMArena Chinese | 1242 | 1493 |
| LMArena French | 1283 | 1461 |
| LMArena German | 1251 | 1439 |
| LMArena Japanese | 1209 | 1392 |
| LMArena Korean | 1184 | 1395 |
| LMArena Russian | 1251 | 1432 |
| LMArena Spanish | 1261 | 1456 |
Instruction Following Hy3 leads
GPT-4: 65.3 (#222), Hy3: 75.1 (#70)
| Benchmark | GPT-4 | Hy3 |
|---|---|---|
| LMArena Instruction Following | 1241 | 1426 |
Long Context Hy3 leads
GPT-4: 37.7 (#212), Hy3: 44.1 (#75)
| Benchmark | GPT-4 | Hy3 |
|---|---|---|
| LMArena Longer Query | 1244 | 1442 |
Writing & Preference Hy3 leads
GPT-4: 34.9 (#268), Hy3: 62.2 (#81)
| Benchmark | GPT-4 | Hy3 |
|---|---|---|
| LMArena Text | 1263 | 1439 |
| LMArena Creative Writing | 1244 | 1402 |
| LMArena Multi-Turn | 1257 | 1436 |
| EQ-Bench Creative Writing | 752 | — |
Frequently asked questions
Is GPT-4 better than Hy3?
Hy3 is the stronger model overall, scoring 44.2 to 29.1 on the Noometry Index.
Which is cheaper, GPT-4 or Hy3?
Hy3 is cheaper. It lists at $0.0825 per million input tokens and $0.33 per million output tokens; GPT-4 lists at $30 and $60.
Is GPT-4 or Hy3 better for coding?
Hy3 scores higher on coding benchmarks: 46.8 versus 31.6 in the Noometry coding category.
Which has the bigger context window?
Hy3 does, with 262K tokens against 8K.
How many benchmarks do GPT-4 and Hy3 share?
17 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Hy3 has 19.