MODEL REFERENCE / COHERE

Command A Reasoning
verified model reference.

Cohere reasoning model for multilingual enterprise problem solving, tool use, RAG and agentic workflows.

256,000context tokens
32,000max output
2026-10-06last verified

PRIMARY-SOURCE VERIFIED

Command A Reasoning at a glance.

Cohere documents a 256,000-token context window. Input modalities: text. Output: text. Pricing status: Free preview.

MODEL IDcommand-a-reasoning-08-2025
CONTEXT WINDOW256,000tokens
MAX OUTPUT32,000
KNOWLEDGE CUTOFF2024-06-01
STANDARD RATENot publishedper 1 million tokens
OUTPUT RATE—per 1 million tokens

CAPABILITY CONTRACT

Published model behavior.

REASONING

configurable

Only provider-documented reasoning controls are shown; missing controls remain unpublished rather than inferred.

INPUT

text

Input modalities come from the cited official model documentation.

OUTPUT

text

Output modality and output-limit claims remain separate so an undocumented token cap is not guessed.

PRICING STATUS

Free preview.

Free preview.

SXF preserves the provider-published native billing basis. Only calculator-eligible generative token models enter token-cost arithmetic; specialist units and input-only embeddings remain visible without forced conversion.

CAVEATS

What the source record does not assume.

01

Cohere currently lists command-a-reasoning-08-2025 as Live with a 256K context window and 32K maximum output.

02

Cohere states that both trial and production API keys are free until rate limits are reached; no numeric pay-as-you-go token rate is asserted.

03

The model is excluded from the calculator because no paid Standard token rate is published for the Cohere API.

EXPLORE

Keep the model in context.

OFFICIAL SOURCES

Verification links.

WHAT CHANGED

Command A Reasoning: what changed.

No post-baseline factual changes recorded for Command A Reasoning since 2026-10-06.

VERIFIED BASELINE2026-10-06Current facts remain aligned with the SXF ledger.

VERIFIED CHANGE HISTORY

Command A Reasoning verified history.

Open ledger JSON ↗
BASELINE

Command A Reasoning

Baseline verified: 256,000 context · 32,000 max output

Official evidence ↗

Append-only SXF ledger · 1 verified event · chain head a81376cb9fe2…