Other●●Solid
I built a live race between LLMs to see which generates tokens fastest
Live drag race UI, but latency benchmarks are API-dependent, not architecturally novel.
Eye Candy
baristaGeek
114mo ago
Back to browse
GitHub Repository
Measure tokens/sec of any LLM behind an OpenAI-compatible API.
1 starsGo
I built a token-per-second counter in Go
by canergl·Jul 23, 2026·1 point·0 comments
AI Analysis
●●SolidShip ItNiche Gem
Handy CLI for benchmarking TPS, but curl scripts already do this for free.
Strengths
- •Built-in warmup runs and median calculation prevent cold-start skew.
- •Works with any OpenAI-compatible endpoint, including local servers.
- •Single static binary via go install makes deployment trivial.
Weaknesses
- •Functionality is easily replicable with a simple curl and awk script.
- •Lacks persistent history or comparison features for tracking trends.
Category
Target Audience
LLM developers and engineers benchmarking model providers
Similar To
llm-bench · custom curl scripts · Artillery
Similar Projects
Developer Tools●Mid
OpenCode Benchmark Dashboard
Benchmarks OpenCode models locally, but lacks preloaded datasets and only works with configured OpenAI-compatible APIs.
Niche Gem
grigio
104mo ago
Developer Tools●Mid
Iamspeed.dev – Fast.com Style but for LLMs
Fast.com for LLMs, but you could script this with curl in five minutes.
Cozy
qainsights
201mo ago
Developer Tools●●Solid
In-browser real LLM token counter and cost estimation
Local tokenizer execution beats sending text to external counters.
CozySolve My Problem
angelmm
311mo ago
AI/ML●Mid
Benchmark multiple LLMs to compare quality, speed, and cost
Yet another prompt benchmarking UI when Promptfoo and LangSmith already exist.
SlickShip It
henriklipp
303mo ago
AI/ML●●Solid
Kitchen Rush, Overcooked inspired LLM tool calling benchmark
Thinking time becomes game time lost — finally measures latency alongside accuracy.
Big BrainNiche Gem
bombastic311
301mo ago