llm.ing

MiniMax-M3.1-Flash

MiniMax (minimax.io)open weightsOpenRouter ↗Hugging Face ↗

Frontier multimodal coding model with 1M context window and tunable thinking depth

Benchmarks

BenchmarkCategoryScoreVariantProvenanceSourceObserved
AA Intelligence Indexintelligence29.1—mirroropenrouterNov 4, 2026
LMArena WebDevpreference1584—crowdlmarenaNov 11, 2026

Includes index data from Artificial Analysis.

Providers

Live operational stats per hosting endpoint, via OpenRouter.

ProviderQuantContext$/1M in · outUptime 24hLatencyThroughput
DigitalOcean—1.0M$0.15 · $0.6100.00%——
Fireworks—1.0M$0.15 · $0.699.31%——
GMICloudfp81.0M$0.15 · $0.6100.00%——
Inferact vLLMnvfp41.0M$0.15 · $0.6100.00%——
InferenceNetfp41M$0.15 · $0.698.62%——
InferenceNetfp41M$0.15 · $0.698.61%——
Makorafp41.0M$0.15 · $0.599.93%——
Minimax—1.0M$0.075 · $0.399.03%——
ModelRunfp41.0M$0.15 · $0.6100.00%——
Nebiusfp41.0M$0.15 · $0.6100.00%——
Parasailfp81.0M$0.15 · $0.6100.00%——
SambaNova—1.0M$0.06 · $0.2499.80%——
StreamLakefp81M$0.135 · $0.54100.00%——
Together—1.0M$0.15 · $0.6100.00%——
Wafernvfp41.0M$0.135 · $0.5499.93%——