Aider Polyglot
Harbor nativeAider Polyglot: 6种语言diff编辑
Verified71%Coding
- Samples
- 225
- Released
- 2024
- Scores
- 61
LMArena
LMArena (Chatbot Arena) Text Leaderboard
LMArena: 众包真实用户匿名对战，Bradley-Terry/Elo 评分
How to run
ltb · lmarena
$ npm i -g lettokenburn-cli$ ltb bench info lmarena
External harness. No Harbor adapter has been generated for this benchmark yet. `ltb bench info` prints the known links.
Runs happen on your machine with your own subscription; only the score summary is pushed. CLI docs
Scores
17 sourced entries| Harness | Source | Confidence | Link | |||
|---|---|---|---|---|---|---|
| Claude Opus 4.6 | 1503.0 | N/A | Self-reported | low | Jun 1, 2026 | lmarena.ai |
| Gemini 3.1 Pro | 1492.0 | N/A | Self-reported | low | Jun 1, 2026 | lmarena.ai |
| Gemini 3 Pro | 1486.0 | N/A | Self-reported | low | Jun 1, 2026 | lmarena.ai |
| GPT-5.4 Pro | 1480.0 | N/A | Self-reported | low | Jun 1, 2026 | lmarena.ai |
| GPT-5.4 | 1463.0 | N/A | Self-reported | low | Jun 1, 2026 | lmarena.ai |
| Claude Sonnet 4.6 | 1460.0 | N/A | Self-reported | low | Jun 1, 2026 | lmarena.ai |
| GLM-5 | 1454.0 | N/A | Self-reported | low | Jun 1, 2026 | lmarena.ai |
| Qwen 3.5 | 1450.0 | N/A | Self-reported | low | Jun 1, 2026 | lmarena.ai |
| Kimi K2.5 | 1438.0 | N/A | Self-reported | low | Jun 1, 2026 | lmarena.ai |
| DeepSeek V3.2 | 1423.0 | N/A | Self-reported | low | Jun 1, 2026 | lmarena.ai |
| Mistral Large | 1416.0 | N/A | Self-reported | low | Jun 1, 2026 | lmarena.ai |
| Grok 3 | 1412.0 | N/A | Self-reported | low | Jun 1, 2026 | lmarena.ai |
| MiniMax M2.5 | 1404.0 | N/A | Self-reported | low | Jun 1, 2026 | lmarena.ai |
| DeepSeek R1 | 1398.0 | N/A | Self-reported | low | Jun 1, 2026 | lmarena.ai |
| MiMo-V2-Flash | 1393.0 | N/A | Self-reported | low | Jun 1, 2026 | lmarena.ai |
| GPT-oss 120B | 1355.0 | N/A | Self-reported | low | Jun 1, 2026 | lmarena.ai |
| Llama 4 Maverick | 1328.0 | N/A | Self-reported | low | Jun 1, 2026 | lmarena.ai |
Every number links to where it was published. Low-confidence rows are muted and need per-cell verification.
Reproduced by the community
All runsLoading community runs…
About
LMArena (LMArena (Chatbot Arena) Text Leaderboard) — 众包真实用户匿名对战，Bradley-Terry/Elo 评分. Scoring: Elo. Difficulty: 高. Imported from benchmark_research.md (2026-06).