# Prism by Ssimplifi

> Prism is an OpenAI-compatible AI API proxy that prints your savings on the invoice — measurable, multi-provider, drop-in compatible. Three-layer response caching (exact + semantic + provider-native passthrough) + intelligent routing across Anthropic, OpenAI, and Google + session memory + edge serving via Cloudflare Workers + per-project policy and budget governance.

## Documentation

- [API Documentation](https://ssimplifi.com/docs): Complete API reference with curl and Python examples
- [Pricing](https://ssimplifi.com/pricing): Free, Pro ($19/mo), and Team ($49/mo) tiers
- [Getting Started](https://ssimplifi.com/signup): Free tier — 50K tokens/day, no credit card required

## Guides (pillar pages)

- [AI API Caching](https://ssimplifi.com/guides/ai-api-caching): Exact, semantic, and provider-native — the three layers that cut AI bills by half.
- [LLM Cost Reduction](https://ssimplifi.com/guides/llm-cost-reduction): 14 techniques ranked by ROI, with measured savings.
- [OpenAI Cost Optimization](https://ssimplifi.com/guides/openai-cost-optimization): Every technique that actually cuts OpenAI bills.
- [AI Gateway Comparison](https://ssimplifi.com/guides/ai-gateway-comparison): Feature matrix for every major AI gateway in 2026.
- [LLM Observability](https://ssimplifi.com/guides/llm-observability): What to instrument first, what to skip.
- [LLM Budget Governance / AI FinOps](https://ssimplifi.com/guides/llm-budget-governance): AI FinOps for engineering teams.
- [Multi-Region LLM API](https://ssimplifi.com/guides/multi-region-llm-api): Edge inference, cache replication, latency budgets.
- [OpenAI-Compatible API](https://ssimplifi.com/guides/openai-compatible-api): The substrate eating the LLM market.

## Comparisons

- [Prism vs Portkey](https://ssimplifi.com/compare/prism-vs-portkey)
- [Prism vs Helicone](https://ssimplifi.com/compare/prism-vs-helicone)
- [Prism vs LiteLLM](https://ssimplifi.com/compare/prism-vs-litellm)
- [Prism vs OpenRouter](https://ssimplifi.com/compare/prism-vs-openrouter)
- [Prism vs Cloudflare AI Gateway](https://ssimplifi.com/compare/prism-vs-cloudflare-ai-gateway)
- [Prism vs Langfuse](https://ssimplifi.com/compare/prism-vs-langfuse)
- [Prism vs LangSmith](https://ssimplifi.com/compare/prism-vs-langsmith)

## Glossary

- [AI Gateway](https://ssimplifi.com/glossary/ai-gateway)
- [Semantic Cache](https://ssimplifi.com/glossary/semantic-cache)
- [Prompt Caching](https://ssimplifi.com/glossary/prompt-caching)
- [LLM Routing](https://ssimplifi.com/glossary/llm-routing)
- [LLM Observability](https://ssimplifi.com/glossary/llm-observability)
- [AI FinOps](https://ssimplifi.com/glossary/ai-finops)
- [Edge Inference](https://ssimplifi.com/glossary/edge-inference)
- [OpenAI-Compatible Endpoint](https://ssimplifi.com/glossary/openai-compatible-endpoint)
- See `/glossary` for the full 14-term index.

## Free Tools (no email gate)

- [Savings Calculator](https://ssimplifi.com/tools/savings-calculator): Model your monthly savings stack — cache + routing + native passthrough.
- [Cache Hit Rate Estimator](https://ssimplifi.com/tools/cache-hit-rate-estimator): Estimate your cache hit rate from workload shape.
- [Cost Comparison by Model](https://ssimplifi.com/tools/cost-comparison-by-model): Per-token cost across providers.
- [Model Routing Recommender](https://ssimplifi.com/tools/model-routing-recommender): Pick the right routing config.

## Key Features

- Three-layer response caching: exact (Redis fingerprint) + semantic (Upstash Vector + BGE-small embeddings) + provider-native passthrough (Anthropic + OpenAI prompt cache discounts surfaced on the invoice).
- Multi-provider routing across Anthropic Claude, OpenAI GPT, Google Gemini — three quality modes (Eco / Balanced / Sport).
- OpenAI-compatible endpoint at `https://api.ssimplifi.com/v1` — drop-in replacement, no SDK changes.
- Session memory via `X-Prism-Session` header — cross-provider conversation continuity.
- Edge serving via Cloudflare Workers + globally-replicated Workers KV cache.
- Streaming-aware failover + speculative routing (Pro/Team).
- Per-request observability: cost, latency p50/p95/p99, cache status, feedback API.
- Per-project policy + monthly budget caps + audit log (Team tier).
- Multi-payment: Paddle (international) + Razorpay (India).

## Author + Personal Site

- Built by Ravi Patel, solo founder of Ssimplifi.
- Personal blog on LLM infrastructure + AI SaaS: [rikuq.com](https://rikuq.com)