llm.ing

Qwen3-Next 80B-A3B (Thinking)

Alibabaopen weights

Efficient Qwen thinking model for local reasoning, math, and coding agents

Benchmarks

BenchmarkCategoryScoreVariantProvenanceSourceObserved
AA Agentic Indexagentic2.1thinkingmirroropenrouterJul 23, 2026
AA Coding Indexcoding17.4thinkingmirroropenrouterJul 23, 2026
AA Intelligence Indexintelligence11.2thinkingmirroropenrouterOct 5, 2026
LMArena Textpreference1368thinkingcrowdlmarenaNov 9, 2026

Includes index data from Artificial Analysis.

Providers

Live operational stats per hosting endpoint, via OpenRouter.

ProviderQuantContext$/1M in · outUptime 24hLatencyThroughput
Google—262K$0.15 · $1.20.00%——