Rankings
Best for RAG
Models best suited for Retrieval-Augmented Generation workloads.
Models ranked
738
Top context
131K
Llama3 2 11b Vision
Leading provider
Openai
137 models in this ranking
Leaderboard
Tap any row to see full specs and comparisons.
| Rank | Model | Context | Max output | Input $/M | RAG score |
|---|---|---|---|---|---|
| 1 | Llama3 2 11b VisionMeta | 131K | 131K | $0.015/M | 131K ctx |
| 2 | Llama3 2 3bMeta | 131K | 131K | $0.015/M | 131K ctx |
| 3 | gpt-oss-20bOpenai | 131K | 131K | $0.020/M | 131K ctx |
| 4 | Llama 3 2 3bMeta | 131K | 131K | $0.020/M | 131K ctx |
| 5 | Meta Llama 3 1 8b InstructMeta | 131K | 131K | $0.020/M | 131K ctx |
| 6 | Mistral Nemo Instruct 2407Mistral | 131K | 131K | $0.020/M | 131K ctx |
| 7 | Llama Guard 3 8BMeta | 131K | 131K | $0.020/M | 131K ctx |
| 8 | Qwen2 Vl 7bAlibaba | 131K | 131K | $0.020/M | 131K ctx |
| 9 | Llama 3 1 8bMeta | 131K | 16K | $0.020/M | 131K ctx |
| 10 | Meta Llama 3 1 8bMeta | 128K | 2K | $0.020/M | 128K ctx |
| 11 | Hermes3 8bNous Research | 131K | 131K | $0.025/M | 131K ctx |
| 12 | Llama3 1 8bMeta | 128K | 128K | $0.025/M | 128K ctx |
| 13 | Deepseek R1 Distill Llama 8bMeta | 131K | — | $0.025/M | 131K ctx |
| 14 | Qwen3 4b Fp8Alibaba | 128K | 20K | $0.030/M | 128K ctx |
| 15 | Llama 4 Maverick 17b 128e Instruct Fp8Meta | 1M | 16K | $0.050/M | 1000K ctx |
| 16 | Qwen Turbo LatestAlibaba | 1M | 16K | $0.050/M | 1000K ctx |
| 17 | Llama 4 Scout 17b 16eMeta | 10M | 16K | $0.050/M | 10000K ctx |
| 18 | Qwen Turbo 2024 11 01Alibaba | 1M | 8K | $0.050/M | 1000K ctx |
| 19 | Qwen Turbo 2025 04 28Alibaba | 1M | 16K | $0.050/M | 1000K ctx |
| 20 | Amazon Nova MicroAmazon | 128K | 10K | $0.035/M | 128K ctx |
| 21 | Nova Micro 1.0Amazon | 128K | 10K | $0.035/M | 128K ctx |
| 22 | Apac Amazon Nova MicroAmazon | 128K | 10K | $0.037/M | 128K ctx |
| 23 | Qwen3 8b Fp8Alibaba | 128K | 20K | $0.035/M | 128K ctx |
| 24 | Mistral NemoMistral | 131K | 4K | $0.040/M | 131K ctx |
| 25 | Nvidia Nemotron Nano 9bNvidia | 131K | 131K | $0.040/M | 131K ctx |
Showing 25 of 738 models
More rankings
Explore other model leaderboards.
Largest Context Window
Models ranked by maximum context window size.
Best for AI Agents
Models with the capabilities needed to power autonomous agent workflows.
Best for Document Processing
Models with large enough context to process long documents in one pass.
Best Value per Context Token
Models offering the most context window per dollar of input cost.
Best Multimodal
Vision-capable models with the largest context windows.
Best Reasoning
Models with extended thinking or strong reasoning capabilities.
Best for Chatbots
Fast, cost-efficient models ideal for real-time conversational applications.