LM Speed
LLM API Speed Test & Leaderboard
FreeAI253 impressions#1202 of its week
LM Speed lets developers test, compare and audit the speed and cost of LLM APIs across many providers.
- for
- Developers building or optimizing AI applications that rely on LLM APIs.
- pricing
- freemium
Key features
- Model score lookup — Search models to view benchmark scores for coding, reasoning, math and more.
- Pricing comparison — Side-by-side per-token pricing across 100+ providers, including free tiers.
- Real-time speed benchmarks — Run five-round tests measuring latency, throughput and total duration.
- API security audit — Check model authenticity, hidden prompts, instruction tampering and error leakage.
- Custom endpoint testing — Benchmark any OpenAI-compatible URL, API key and model ID.
- Analytics dashboard — Review latency, throughput, health signals and stability rankings.
Use cases
- Select the fastest LLM provider for a production chatbot.
- Find the cheapest per-token pricing for a specific model.
- Audit a third-party proxy for hidden prompts or response tampering.
- Monitor ongoing health and performance of a chosen API.
- Benchmark a self-hosted LLM endpoint before deployment.
LM Speed pricing
- Site verification$0Publish verification code · Standard review after verification
- Paid provider publishing$19.90one-timePriority manual review · No subscription or renewal
LM Speed vs alternatives
 LM Speed |  OpenRouter |  Ollama |  Hugging Face |  Replicate | |
|---|---|---|---|---|---|
| Best for | Speed and cost benchmarking for LLM APIs | Unified model access | Local model hosting | Model discovery | Model execution and fine-tuning |
| Pricing | Freemium | Subscription | Free | Free | Subscription |
| DevHunt upvotes | 0 | 0 | 0 | 0 | 0 |
| Launched | Jan 2026 | — | — | — | — |
- LM Speed vs OpenRouter: Provides a single API to access many models but lacks built-in speed or security benchmarking.
- LM Speed vs Ollama: Runs models locally, so no external API speed testing or pricing comparison.
- LM Speed vs Hugging Face: Model hub and dataset repository without automated speed or security audits.
- LM Speed vs Replicate: Offers API for running/fine-tuning models but does not focus on latency or pricing analytics.
LM Speed FAQ
How do I query model benchmark scores?+
Use the model search on the homepage or the Model Performance leaderboard to find a model’s detailed benchmark page.
How can I compare LLM API pricing across providers?+
LM Speed aggregates per-token pricing from over 100 providers; each model page shows a side-by-side pricing table.
Which LLM APIs are free?+
Many providers offer free tiers or credits for models like DeepSeek, Gemini and Llama; see the Free LLM API directory for details.
How does LM Speed conduct speed benchmark testing?+
It runs a five-round continuous stress test with standardized prompts, measuring first-token latency and output throughput using tiktoken.
What does the API trust audit check?+
It sends safety probes to verify model identity, hidden prompts, instruction rewriting, stream integrity and error leakage, returning a risk score and report.
How to compare speed between different API providers?+
Use the performance leaderboards and model detail pages to compare throughput, latency and health rankings across providers.
Summarized by DevHunt from lmspeed.net · Oct 5, 2026. Details may change; check the official site.
.png?width=64)
.png?width=64)
.png?width=64){kind=logo}
