RunInfra is now public.See what's new
RunInfra
Own your AI. We benchmark GPUs, optimize kernels, and deploy open-source models as production APIs.
Start building
Embeddings and rerank, one round-trip.
BGE, E5, GTE, Nomic. Encoder and cross-encoder reranker fused on a single GPU.
Search
Recs
Classification
Reranking
What you actually own
The optimization knobs, the codebase, the model choice. None of it locked away.
01
Encoder plus reranker, fused.
Cross-encoder reranks the same top-k in the same vLLM server. One round-trip beats the stacked Pinecone-plus-Cohere setup.
02
100+ languages, one recipe.
BGE-M3 covers Chinese, English, French, Hindi, and 96+ more with the same embedding-serving recipe. No model-per-language sprawl.
03
5-10x cheaper at scale.
Closed APIs bill per call. RunInfra amortizes GPU compute across batches. At 1M docs a day, you save thousands a month.
Three ways to ship embeddings
Most teams pick between speed and control. RunInfra keeps both in one workflow.
| What matters | RunInfraRecommendedFast path with model control and export. | Closed embedding APIsPer-call, hosted. | DIY self-hostingFull control, heavy operations. |
|---|---|---|---|
| 01Launch | Pick model, optimize, deploy Start quickly and keep the production path open. | Call provider endpoint Fast first demo, but the runtime stays rented. | Build serving stack first Infrastructure work comes before product learning. |
| 02Model control | Bring the model ID Keep model choice and serving decisions visible. | Provider catalog You use what the provider exposes. | Your model Full control if your team maintains the runtime. |
| 03Tuning | Measured latency and GPU cost Compare serving choices before deployment. | Opaque Latency and batching stay behind the API. | Manual profiling Your team owns tuning and regressions. |
| 04Export | Managed now, export when needed Use the endpoint first and take the deploy package later. | Locked endpoint You keep calling the provider. | Already owned Export exists because you built everything yourself. |
| 05Operations | Low until you choose to own it Operate managed, then export with the same measured plan. | Low, with lock-in Less infra work, less production control. | High You own infra, failures, upgrades, and serving changes. |
| 06Security | SOC 2 Type 2 Audited controls across access, logging, and incident response. | Varies by vendor Compliance depends on the third party sitting in the request path. | You build it Your team owns the audit trail, logging, and access controls. |
RunInfra
RecommendedFast path with model control and export.
Launch
Pick model, optimize, deploy
Start quickly and keep the production path open.
Model control
Bring the model ID
Keep model choice and serving decisions visible.
Tuning
Measured latency and GPU cost
Compare serving choices before deployment.
Export
Managed now, export when needed
Use the endpoint first and take the deploy package later.
Operations
Low until you choose to own it
Operate managed, then export with the same measured plan.
Security
SOC 2 Type 2
Audited controls across access, logging, and incident response.
Code you own. Deploy anywhere.
The full recipe ships with you. Codebase, kernels, engine config, weights. Run it anywhere.
Live
Switch to E5 Mistral and rerun the bench...
Vector adapters3
runinfra-embeddings/
models/
evals/
Managed RunInfra
Our GPUs, per-million-tokens billing from L4 to B200.
Your infrastructure
AWS, GCP, RunPod, bare metal. Same Dockerfile, your cluster.
Local workstation
docker compose up. Full pipeline on a single GPU.
Search the Hugging Face catalog
Encoders and rerankers, live from huggingface.co. Click through to inspect, or paste any compatible ID into the dashboard.
HF
What RunInfra tunes
Every stage of the pipeline, retuned per model and GPU.
Batch indexing
Continuous batching for offline indexing. 12k tokens/s peak on L4.
Streaming embed
Low-latency single-query embedding for live search. p50 4ms.
Reranker fusion
Cross-encoder reranker in the same server. One round-trip for top-k.
Embedding serving sweep
Tune batch size, token budget, and padding policy against recall and cosine gates.
Vector DB sync
Push to Pinecone, Qdrant, Weaviate, or pgvector. Built-in sink.
Per-stage scheduling
Encode plus rerank interleaved. Late-interaction pooling on retrieval.
Try this pipeline
Edit the model, engine, or GPU inline. Send to retune the stack in the dashboard.
Common questions
Can't find what you're looking for? Get in touch
Which embedding models work?
Any Hugging Face encoder that loads through vLLM or sentence-transformers. BGE family is the recommended starting point for English, BGE-M3 for multilingual.
