llm.ing

Llama-3.3-70B-Instruct

Metaopen weightsOpenRouter ↗Hugging Face ↗

Open Llama instruction model for multilingual chat, reasoning, and coding

Benchmarks

BenchmarkCategoryScoreVariantProvenanceSourceObserved
AA Agentic Indexagentic0.3—mirroropenrouterJul 23, 2026
AA Coding Indexcoding11.9—mirroropenrouterJul 23, 2026
AA Intelligence Indexintelligence7.7—mirroropenrouterOct 5, 2026
LMArena Textpreference1274—crowdlmarenaNov 9, 2026

Includes index data from Artificial Analysis.

Providers

Live operational stats per hosting endpoint, via OpenRouter.

ProviderQuantContext$/1M in · outUptime 24hLatencyThroughput
AkashMLfp8131K$0.2 · $0.5298.44%——
Cloudflarefp824K$0.293 · $2.2596.21%——
CoreWeavefp16128K$0.71 · $0.7199.96%——
Crusoebf16131K$0.25 · $0.7599.51%——
DeepInfrafp8131K$0.1 · $0.3298.41%——
Google—128K$0.72 · $0.7298.86%——
Google—128K$0.72 · $0.7299.96%——
Groq—131K$0.59 · $0.7999.83%——
Nebiusfp8131K$0.13 · $0.440.86%——
Novitabf1612K$0.135 · $0.495.65%——
Parasailfp8131K$0.22 · $0.599.39%——
SambaNova—131K$0.45 · $0.999.12%——
Together—131K$1.04 · $1.0498.27%——