AI Model Radar: Sources and Selection Guide
Compare model candidates using traceable sources and your own workloads. Historical ranking data below has not been independently reverified; it is not a current leaderboard.
Review date: 2026-10-11. Source collection date: unknown. The review date does not certify the model facts.
How to compare candidates
- Use the exact same prompts and record the model ID, API protocol and generation settings.
- For coding, check executable results and tool-call arguments. A model name alone does not establish protocol support.
- For documents, check retrieval accuracy and citations at the actual input size.
- Record total billed cost, latency and failed requests. Compare several runs; one example is not a benchmark.
Filter models
20 model candidates
Availability, pricing and benchmark scores require a current check. Missing evidence does not mean a capability is unsupported.
Interactive data could not load. The static model list remains available below.
| Model | Provider | Source to verify | Evidence status |
|---|---|---|---|
| Claude 3.7 Sonnet | Anthropic | LMSYS Arena · Artificial Analysis · OpenCompass | Not verified |
| GPT-4o / GPT-4o mini | OpenAI | LMSYS Arena · Artificial Analysis · OpenCompass | Not verified |
| Gemini 2.5 Pro | LMSYS Arena · Artificial Analysis · OpenCompass | Not verified | |
| Gemini Flash family | LMSYS Arena · Artificial Analysis · OpenCompass | Not verified | |
| DeepSeek R1 | DeepSeek | LMSYS Arena · Artificial Analysis · OpenCompass | Not verified |
| DeepSeek V3 / V3.2 | DeepSeek | OpenRouter Rankings · LMSYS Arena · Artificial Analysis · OpenCompass | Not verified |
| Claude Sonnet 4.6 | Anthropic | OpenRouter Rankings · LMSYS Arena · Artificial Analysis · OpenCompass | Not verified |
| Claude Opus 4.6 | Anthropic | OpenRouter Rankings · LMSYS Arena · Artificial Analysis · OpenCompass | Not verified |
| Grok 4 | xAI | LMSYS Arena · Artificial Analysis · OpenCompass | Not verified |
| Qwen3-Max | Qwen | LMSYS Arena · Artificial Analysis · OpenCompass | Not verified |
| Llama 4 Maverick | Meta | LMSYS Arena · Artificial Analysis · OpenCompass | Not verified |
| Mistral Medium 3 | Mistral | LMSYS Arena · Artificial Analysis · OpenCompass | Not verified |
| Gemini 3.1 Pro | OpenRouter Rankings · LMSYS Arena · Artificial Analysis · OpenCompass | Not verified | |
| GPT-5.4 | OpenAI | LMSYS Arena · Artificial Analysis · OpenCompass | Not verified |
| Stepfun / Step 3.5 Flash | Stepfun | OpenRouter Rankings | Not verified |
| GLM-5 Turbo | Z.ai | OpenRouter Rankings | Not verified |
| Xiaomi MiMo V2 Pro | Xiaomi | OpenRouter Rankings | Not verified |
| OpenRouter Hunter Alpha | OpenRouter | OpenRouter Rankings | Not verified |
| OpenRouter Optimus Alpha | OpenRouter | Not verified | Not verified |
| Minimax M2 5 20260211 | MiniMax | OpenRouter Rankings | Not verified |
No matching models.
Sources and verification
The stored dataset combines OpenRouter usage rankings and manually maintained Arena, Artificial Analysis and OpenCompass entries. Collection dates and per-model evidence are missing, so historical scores and rankings are not displayed as current facts. Visit the linked sources to verify exact model variants and dates.
Review date: 2026-10-11. Source collection date: unknown. The review date does not certify the model facts.