googleimage + text + video → text
Google: Gemma 4 26B A4B
Canonical ID: google/gemma-4-26b-a4b-it
Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind. Despite 25.2B total parameters, only 3.8B activate per token during inference — delivering near-31B quality at...
Artificial Analysis Benchmarks
VerifiedStrategy & Ops5.3 / 100
Finance & Accounting7.1 / 100
Legal Index6.9 / 100
Token Pricing & Economics
Prompt Input (1M Tokens)
Input$0.07
Completion Output (1M Tokens)
Output$0.22
Prompt KV-Cache Read (1M Tokens)
Cached$0.04
Context Window Capacity
Context256k tokens
Supported Parameters & Features
frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_completion_tokensmax_tokensmin_ppresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p