Model comparison
Hy3 vs o3
o3 is the stronger model overall, scoring 47.4 to 43.6 on the Noometry Index. Hy3 costs 15× less per token, which makes it the better buy when o3's lead doesn't matter for your workload.
Last verified . 15 shared benchmarks.
TencentSummary
- They share 15 benchmarks with published results for both. Hy3 scores higher in 2 categories and o3 in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where o3 leads 54.5 to 40.2.
- Hy3 is cheaper at $0.13 / $0.53 per million input/output tokens, against $2 / $8 for o3.
- Hy3 accepts more context: 262K tokens versus 200K.
- Hy3 has downloadable open weights; the other is API-only.
Side by side
| Hy3 | o3 | |
|---|---|---|
| Provider | Tencent | OpenAI |
| Noometry Index | 43.6 | 47.4 |
| Released | 2026-07-06 | 2025-04-16 |
| Weights | Open | Proprietary |
| Context window | 262K | 200K |
| Max output | 128K | 100K |
| Input $ / M tokens | $0.13 | $2 |
| Output $ / M tokens | $0.53 | $8 |
| Results tracked | 17 | 63 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Hy3: 46.6 (#60), o3: 47.3 (#57)
| Benchmark | Hy3 | o3 |
|---|---|---|
| LMArena Coding | 1465 | 1408 |
| SWE-bench Verified | — | 62.3% |
| SWE-bench Verified (bash only) | — | 58.4% |
| Aider Polyglot | — | 81.3% |
| LMArena WebDev | 1509 | — |
| GSO | — | 8.8% |
| WeirdML | — | 52.4% |
| CadEval | — | 74% |
| ALE-Bench | — | 933.55 |
Agentic & Tool Use Not comparable
Hy3: —, o3: 34.3 (#42)
| Benchmark | Hy3 | o3 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 63% |
| GDPval | — | 30.8% |
| DeepResearch Bench | — | 45.2% |
| OSWorld | — | 23% |
| LMArena Search | — | 1144 |
| METR Time Horizons | — | 65.4% |
Reasoning o3 leads
Hy3: 25.0 (#134), o3: 32.1 (#66)
| Benchmark | Hy3 | o3 |
|---|---|---|
| LMArena Hard Prompts | 1449 | 1402 |
| ARC-AGI-2 | — | 6.5% |
| SimpleBench | — | 53.1% |
| Kagi LLM Benchmark | — | 67.6% |
| NYT Connections (extended) | 41.2% | — |
| ARC-AGI-1 | — | 60.8% |
| CritPt | — | 1.4% |
| Chess Puzzles | — | 38% |
| EnigmaEval | — | 13.1% |
| Mystery Game Puzzles | — | 29% |
| DTBench | — | 84.8% |
| LMCA | — | 39.7% |
| Epoch Capabilities Index | — | 146.88 |
| ForecastBench | — | 62.5 |
Math o3 leads
Hy3: 37.4 (#98), o3: 48.9 (#49)
| Benchmark | Hy3 | o3 |
|---|---|---|
| LMArena Math | 1475 | 1426 |
| FrontierMath (Tiers 1-3) | — | 33.3% |
| OTIS Mock AIME 2024-2025 | — | 84.4% |
| Omni-MATH | — | 71.4% |
| MATH Level 5 | — | 97.8% |
| FrontierMath (Feb 2025 set) | — | 18.7% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge o3 leads
Hy3: 40.2 (#108), o3: 54.5 (#47)
| Benchmark | Hy3 | o3 |
|---|---|---|
| LMArena Expert | 1461 | 1403 |
| GPQA Diamond | — | 81.8% |
| Humanity's Last Exam | — | 20.3% |
| SimpleQA Verified | — | 49.4% |
| MMLU-Pro | — | 85.9% |
| Confabulations | — | 14.4% |
| GPQA (HELM) | — | 75.3% |
Multimodal Not comparable
Hy3: —, o3: 40.7 (#35)
| Benchmark | Hy3 | o3 |
|---|---|---|
| LMArena Vision | — | 1214 |
| GeoBench | — | 74% |
| VPCT | — | 52% |
Multilingual Hy3 leads
Hy3: 53.7 (#60), o3: 51.9 (#99)
| Benchmark | Hy3 | o3 |
|---|---|---|
| LMArena Non-English | 1427 | 1402 |
| LMArena Chinese | 1496 | 1438 |
| LMArena French | 1461 | 1430 |
| LMArena Korean | 1390 | 1371 |
| LMArena Russian | 1435 | 1407 |
| LMArena Spanish | 1448 | 1395 |
| LMArena German | — | 1420 |
| LMArena Japanese | — | 1404 |
Instruction Following Hy3 leads
Hy3: 75.3 (#68), o3: 73.0 (#122)
| Benchmark | Hy3 | o3 |
|---|---|---|
| LMArena Instruction Following | 1425 | 1368 |
| IFEval | — | 86.9% |
Long Context o3 leads
Hy3: 44.1 (#70), o3: 53.2 (#6)
| Benchmark | Hy3 | o3 |
|---|---|---|
| LMArena Longer Query | 1443 | 1372 |
| Fiction.LiveBench | — | 88.9% |
| CL-bench | — | 17.8% |
Writing & Preference o3 leads
Hy3: 62.5 (#75), o3: 63.7 (#62)
| Benchmark | Hy3 | o3 |
|---|---|---|
| LMArena Text | 1442 | 1410 |
| LMArena Creative Writing | 1399 | 1359 |
| LMArena Multi-Turn | 1436 | 1405 |
| Short-Story Creative Writing | — | 83.9% |
| EQ-Bench Creative Writing | — | 1676 |
| WildBench | — | 86.1% |
Frequently asked questions
Is Hy3 better than o3?
o3 is the stronger model overall, scoring 47.4 to 43.6 on the Noometry Index. Hy3 costs 15× less per token, which makes it the better buy when o3's lead doesn't matter for your workload.
Which is cheaper, Hy3 or o3?
Hy3 is cheaper. It lists at $0.13 per million input tokens and $0.53 per million output tokens; o3 lists at $2 and $8.
Is Hy3 or o3 better for coding?
They score almost the same on coding (46.6 vs 47.3); test both on your own repository before choosing.
Which has the bigger context window?
Hy3 does, with 262K tokens against 200K.
How many benchmarks do Hy3 and o3 share?
15 benchmarks have published results for both models. Hy3 has 17 scored results on Noometry and o3 has 63.