Open-source dataset tracking LLMOps platforms, LLM inference engines, orchestration frameworks, prompt management tools, and observability solutions. Compare 80+ tools. Updated weekly.
| Tool | Category | Throughput (tok/s) | License | Deployment |
|---|---|---|---|---|
| vLLM | Inference Engine | 3,200 | Apache 2.0 | Self-hosted |
| TGI (HuggingFace) | Inference Server | 2,800 | Apache 2.0 | Self-hosted/Cloud |
| LangChain | Orchestration | N/A | MIT | Any |
| Langfuse | Observability | N/A | MIT | SaaS/Self-hosted |
| Ollama | Local Inference | 450 | MIT | Local |
Deploy vLLM, TGI, and Ollama on RunPod GPUs. H100 and A100 instances for high-throughput LLM serving.
Try RunPod →On-demand and reserved H100 instances optimized for LLM training and inference workloads.
Try Lambda →GPU marketplace for cost-effective LLM inference. Find the best price-to-performance ratio.
Try Vast.ai →