Leading: Sendbird (60.9%)
| Metric | Value |
|---|---|
| NATS wins | 9 |
| Sendbird wins | 14 |
| Abstains (no tool) | 41 |
| Other tool chosen | 3502 |
| Decisive cases | 23 |
| NATS win rate (unweighted) | 39.1% |
| 95% CI | 22.2% - 59.2% |
| NATS win rate (weighted) | 39.1% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model | Tier | NATS | Sendbird | None | Other | A rate |
|---|---|---|---|---|---|---|
| Gemini 2.5 Pro | Frontier | 0 | 13 | 0 | 176 | 0% |
| Claude Opus 4.6 | Frontier | 9 | 0 | 0 | 179 | 100% |
| Llama 4 Maverick | Frontier | 0 | 1 | 0 | 188 | 0% |
| Claude Haiku 4.5 | Small | 0 | 0 | 0 | 184 | n/a |
| Claude Sonnet 4.6 | Frontier | 0 | 0 | 0 | 189 | n/a |
| DeepSeek R1 0528 | Frontier | 0 | 0 | 30 | 158 | n/a |
| DeepSeek V3.2 | Mid | 0 | 0 | 1 | 188 | n/a |
| Devstral 2 2512 | Mid | 0 | 0 | 0 | 189 | n/a |
| Gemini 2.5 Flash | Small | 0 | 0 | 0 | 189 | n/a |
| GLM 5 Turbo | Frontier | 0 | 0 | 3 | 186 | n/a |
| GPT 5.3 Codex | Frontier | 0 | 0 | 0 | 189 | n/a |
| GPT 5.4 | Frontier | 0 | 0 | 0 | 186 | n/a |
| GPT 5.4 Mini | Mid | 0 | 0 | 2 | 183 | n/a |
| Kimi K2.5 | Frontier | 0 | 0 | 5 | 182 | n/a |
| Llama 4 Scout | Small | 0 | 0 | 0 | 186 | n/a |
| MiMo V2 Pro | Frontier | 0 | 0 | 0 | 188 | n/a |
| MiniMax M2.7 | Frontier | 0 | 0 | 0 | 187 | n/a |
| Mistral Small 4 | Mid | 0 | 0 | 0 | 187 | n/a |
| Qwen3 Coder Next | Mid | 0 | 0 | 0 | 188 | n/a |
| Prompt | Tier | NATS | Sendbird | None | Other | A rate |
|---|---|---|---|---|---|---|
| chat-application | Beginner | 0 | 12 | 3 | 382 | 0% |
| social-media-platform | Advanced | 6 | 0 | 3 | 389 | 100% |
| chat-application | Advanced | 3 | 1 | 1 | 390 | 75% |
| chat-application | Intermediate | 0 | 1 | 6 | 390 | 0% |
| project-management-tool | Beginner | 0 | 0 | 4 | 394 | n/a |
| project-management-tool | Intermediate | 0 | 0 | 3 | 393 | n/a |
| project-management-tool | Advanced | 0 | 0 | 0 | 398 | n/a |
| social-media-platform | Beginner | 0 | 0 | 17 | 374 | n/a |
| social-media-platform | Intermediate | 0 | 0 | 4 | 392 | n/a |