AI Model Timeline

Tracking the accelerating release frequency of frontier AI models.

Command A

Cohere·

LLMsopen-weightcloud + localWorkstation

Overview

111B Dense Enterprise-Flagship (command-a-03-2025), 256k Kontext, 23 Sprachen, Fokus auf Tool-Use, agentische Workflows und Translation. Läuft auf wenigen GPUs, offene Gewichte. Direkter Vorgänger von Command A+. Vectara HHEM 9.3% Halluzination.

Capabilities and innovations

111B DenseEnterprise / Tool-Use / Agentic23 Sprachen256k ContextOpen weightsEffizientes Enterprise-FlagshipOpen weights

Benchmarks

BenchmarkScoreSource
GPQA Diamond52.7%vendor
Command A Technical Report (simple-evals): GPQA Diamond 52.7, MMLU-Pro 71.2.
MMLU-Pro71.2%vendor
Command A Technical Report (simple-evals).
  • Humanity’s Last Exam (no tools): not reported by the vendor
  • SWE-bench Verified: not reported by the vendor
  • SWE-bench Pro: benchmark did not exist at release
  • Terminal-Bench 2.x: benchmark did not exist at release
  • MMMU-Pro: benchmark did not exist at release

Architecture and hardware

Parameters
111B total, 111B active per token (Dense)
Estimated VRAM at Q4
~64 GB, Workstation class

Links

More from Cohere

Data curated by AI Model Timeline. See the methodology for admission criteria, benchmark eras and source priority.