PRODUCT B • MODELS HUBLive Leaderboard
OpenRely Models Hub & LLM Benchmarks
Aggregating DeepSeek, Claude, Gemini, and GPT series models. Providing real-time TPS throughput, TTFT latency benchmarks, transparent pricing, and tailored recommendations.
Real-time TPS Benchmarks
Measure real-time tokens per second and time-to-first-token across global LLM providers.
Transparent Price Benchmark
Compare official rates side-by-side and enjoy up to 50% savings via OpenRely aggregation.
Tailored Model Rankings
Ranked by code generation, reasoning, long-context, and conversational capabilities.
Live LLM Performance Leaderboard
Updated every 5 minutes
| Rank | Model | Provider | TPS (Throughput) | TTFT (Latency) | OpenRely Price |
|---|---|---|---|---|---|
| #1 | DeepSeek-V4 Pro | DeepSeek | 135.2 tps | 180 ms | $0.14 / 1M tokens |
| #2 | DeepSeek-V4 Flash | DeepSeek | 180.5 tps | 120 ms | $0.07 / 1M tokens |
| #3 | Claude 3.5 Sonnet | Anthropic | 98.4 tps | 240 ms | $3.00 / 1M tokens |
| #4 | Gemini 1.5 Pro | 110.2 tps | 220 ms | $1.25 / 1M tokens | |
| #5 | GPT-4o | OpenAI | 92.1 tps | 260 ms | $2.50 / 1M tokens |
Explore the Standalone Models Hub Site
Visit models.openrely.ai for side-by-side prompt testing, detailed price brackets, and full model specs.