SpacetimeDB LLM Benchmark (C#): leaderboard
C# subset of the SpacetimeDB LLM Benchmark. Tests LLM ability to generate correct SpacetimeDB C# module code, evaluating basics (reducers, tables, CRUD) and schema patterns against live module execution.
Metric: Eval Pass Rate (%). Source: spacetimedb.com. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.6 | 98.9 |
| 2 | Claude Fable 5 | 98.9 |
| 3 | Gemini 3.1 Pro (Preview) | 97.8 |
| 4 | GPT-5.5 | 96.6 |
| 5 | Claude Opus 4.8 | 96.6 |
| 6 | Grok Build 0.1 | 95.5 |
| 7 | Kimi K3 | 93.3 |
| 8 | Grok 4.3 | 93.3 |
| 9 | GPT-5.6 Sol | 92.2 |
| 10 | DeepSeek V4 Flash | 91 |
| 11 | GPT-5.4 Mini | 88.8 |
| 12 | Claude Opus 5 | 66.7 |
| 13 | DeepSeek V4 Pro | 63.4 |
| 14 | Gemini 3.5 Flash | 50.6 |
Interactive version: theaggregate.ai/benchmark?slug=spacetimedb-llm-benchmark-c · How It Works · Data refreshed daily, snapshot 2026-09-05.