Best AI Models for Roo Code
Roo Code is a VS Code extension for AI-assisted development with multiple modes. It supports various models and benefits from strong code generation and instruction following.
What Matters for Roo Code
Best Models for Roo Code
Top 15 by tool-optimized score
All Models Ranked for Roo Code (431 models)
Scored by: benchmark performance (90%) from MMLU, GPQA, HumanEval, SWE-bench, and 15+ standardized evaluations, with capabilities and context as tiebreakers (10%).
| # | Model | Score | Output $/M |
|---|---|---|---|
| 1 | Muse Spark 1.1 Arena Elo: 1493 | 87 | $4.25 |
| 2 | Gemini 3.5 Flash Arena Elo: 1477 | 86 | $9.00 |
| 3 | Claude Opus 5.5 | 85 | $20.00 |
| 4 | Claude Fable 5.1 | 85 | $50.00 |
| 5 | Claude Fable 5.1 (batch) | 85 | $25.00 |
| 6 | GLM 5.3 Flash Arena Elo: 1475 | 85 | $0.500 |
| 7 | Claude Opus 5 | 85 | $25.00 |
| 8 | Grok 4.5 Arena Elo: 1468 | 85 | $6.00 |
| 9 | Claude Fable 5 SWE-bench: 95% | 85 | $50.00 |
| 10 | Claude Fable 5 (batch) | 85 | $25.00 |
| 11 | Claude Opus 4.8 (batch) | 85 | $12.50 |
| 12 | MiMo-V2.5-Pro Arena Elo: 1467 | 85 | $0.870 |
| 13 | Claude Opus 4.7 (batch) | 85 | $12.50 |
| 14 | GLM 5.1 Arena Elo: 1466 | 85 | $3.04 |
| 15 | Gemini 3.5 Flash Lite Arena Elo: 1456 | 84 | $2.50 |
| 16 | Hy3 Arena Elo: 1456 | 84 | $0.330 |
| 17 | Qwen3.7 Plus Arena Elo: 1456 | 84 | $1.28 |
| 18 | Qwen3.6 Max Preview Arena Elo: 1460 | 84 | $6.16 |
| 19 | GPT-5.5 Pro (batch) | 84 | $90.00 |
| 20 | GPT-5.5 (batch) | 84 | $15.00 |
| 21 | Kimi K2.6 Arena Elo: 1460 | 84 | $4.00 |
| 22 | Gemini 3.1 Pro Preview Custom Tools | 84 | $12.00 |
| 23 | Gemini 3.1 Pro Preview (batch) | 84 | $6.00 |
| 24 | GLM 5 Arena Elo: 1458 | 84 | $1.92 |
| 25 | MiniMax M3 Arena Elo: 1441 | 83 | $1.20 |
| 26 | Gemma 4 31B Arena Elo: 1451 | 83 | $0.340 |
| 27 | Qwen3.6 Plus Arena Elo: 1443 | 83 | $1.95 |
| 28 | GPT-5.4 Pro | 83 | $180.00 |
| 29 | GPT-5.4 Pro (batch) | 83 | $90.00 |
| 30 | GPT-5.4 (batch) | 83 | $7.50 |
| 31 | GPT-5.3-Codex | 83 | $14.00 |
| 32 | Qwen3.5 397B A17B Arena Elo: 1442 | 83 | $3.50 |
| 33 | Claude Opus 4.6 (batch) | 83 | $12.50 |
| 34 | GPT-5.2-Codex | 83 | $14.00 |
| 35 | GLM 4.7 Arena Elo: 1442 | 83 | $1.75 |
| 36 | GPT-5.2 Pro | 83 | $168.00 |
| 37 | GPT-5.2 Pro (batch) | 83 | $84.00 |
| 38 | GPT-5.2 (batch) | 83 | $7.00 |
| 39 | Grok 4.7 | 82 | $4.80 |
| 40 | Qwen3.8 27B Arena Elo: 1437 | 82 | $3.00 |
| 41 | Grok 4.6 | 82 | $6.00 |
| 42 | GPT-5.6 Luna Pro | 82 | $1.20 |
| 43 | GPT-5.6 Luna Pro (batch) | 82 | $0.600 |
| 44 | GPT-5.6 Luna | 82 | $1.20 |
| 45 | GPT-5.6 Luna (batch) | 82 | $0.600 |
| 46 | GPT-5.6 Terra Pro | 82 | $12.00 |
| 47 | GPT-5.6 Terra Pro (batch) | 82 | $6.00 |
| 48 | GPT-5.6 Terra | 82 | $12.00 |
| 49 | GPT-5.6 Terra (batch) | 82 | $6.00 |
| 50 | GPT-5.6 Sol Pro | 82 | $10.00 |
More Tool Rankings
Based on our analysis of coding benchmarks, capability matching, and pricing, Muse Spark 1.1 currently ranks #1 for Roo Code. Rankings are rebuilt as benchmark, pricing, and provider data refresh.
We score models using benchmark performance (90%) from LMArena, HumanEval, SWE-bench, MMLU, and 15+ standardized evaluations. Capabilities and context serve as tiebreakers (10%). Only models with the capabilities Roo Code needs are included in the tool-specific rankings.
We currently track 431 AI models compatible with Roo Code. This includes models from OpenAI, Anthropic, Google, DeepSeek, and other providers accessible via API.
Many open-source models are compatible with Roo Code through API providers like OpenRouter, Together AI, and Groq. Check our rankings to see which open-source models perform best.
Rankings refresh whenever the underlying benchmark, pricing, and catalog sources refresh. That means some signals update faster than others, and the page reflects the latest verified source data available.