Real data comparison
OpenAI: GPT-5.6 Luna (batch) vs Tencent: Hy3
On price, OpenAI: GPT-5.6 Luna (batch) is cheaper ($0.22 vs $0.23 per 1M). In real production usage, Tencent: Hy3 is used more (rank #3 vs #9 of 291).
- OpenAI: GPT-5.6 Luna (batch)
- Tencent: Hy3
Head-to-head by real task
OpenAI: GPT-5.6 Luna (batch) leads 8 of 16 shared task categories by real classified traffic share; Tencent: Hy3 leads 8. Not a benchmark score - this is what developers actually route to each model for, per OpenRouter's task classification.
| Task | OpenAI: GPT-5.6 Luna (batch) | Tencent: Hy3 |
|---|---|---|
| File I/O | 7.2% | 13.7% |
| Workflow Execution | 5.7% | 10.4% |
| Code Generation | 9.4% | 5.3% |
| DevOps | 5.3% | 8.9% |
| Code Review | 7.2% | 3.6% |
| SQL & Database | 6.2% | 3.1% |
| Frontend & UI | 10.1% | 7.4% |
| Debugging | 6.9% | 4.2% |
| Shell Execution | 3.7% | 6.0% |
| Research & Reports | 4.8% | 6.7% |
| DevOps & Config | 8.1% | 6.3% |
| Repo Scanning | 6.5% | 4.8% |
| Tool Dispatch | 5.8% | 4.6% |
| Conversation | 3.1% | 4.1% |
| Multi-step Planning | 6.2% | 6.7% |
| Web Search | 4.8% | 5.2% |
Usage share over time
- OpenAI: GPT-5.6 Luna (batch)
- Tencent: Hy3
Frequently asked questions
OpenAI: GPT-5.6 Luna (batch) vs Tencent: Hy3: which is better quality?
Quality data is unavailable for one or both models.
OpenAI: GPT-5.6 Luna (batch) vs Tencent: Hy3: which is cheaper?
OpenAI: GPT-5.6 Luna (batch) is cheaper at $0.22/1M vs $0.23/1M.
OpenAI: GPT-5.6 Luna (batch) vs Tencent: Hy3: which is used more?
Tencent: Hy3 has higher real OpenRouter usage share (rank #3 vs #9 of 291).
OpenAI: GPT-5.6 Luna (batch) vs Tencent: Hy3: which wins more real-world task categories?
OpenAI: GPT-5.6 Luna (batch) and Tencent: Hy3 split the 16 shared task categories evenly.
Source: OpenRouter (openrouter.ai/rankings), as of August 9, 2026.
Benchmark scores: Artificial Analysis (artificialanalysis.ai) via OpenRouter (openrouter.ai/rankings).
Methodology: smophy.ai/benchmark/methodology
Token counts originate from each provider's own tokenizer and are not directly comparable across providers.