Packaging change·Dec 1 → Jan 1, 2025·Dec 2024·AI & Machine Learning
Cohere Introduces Command R7B Model, Optimizing Speed and Efficiency at Competitive Pricing
What changed
Limit decrease
Rerank document length threshold decreased from 4096 tokens to 500 tokens (documents longer than this limit are split into chunks)
Plan renamed
Rerank upgraded from version 3 to version 3.5
New plan
Command R7B model introduced - smallest generative model optimized for speed, efficiency, and quality at $0.0375/1M tokens input, $0.15/1M tokens output
Before and after
Dec 1, 2024 vs Jan 1, 2025Pulse.
100%
Change 1 of 4 · capacity decreased
Rerank document length threshold decreased from 4096 tokens to 500 tokens (documents longer than this limit are split into chunks)
Sign in to view the full comparison
See the before/after screenshots and every marked-up change.
More from Cohere
Full history →- Week 41, 2026·improvedCohere cuts Model Vault Embed pricing 25% with new Embed 5 Fast and Pro tiers
- Week 36, 2026·packagingCohere adds Parse 5 doc-parsing model, Medium $4/hr and XL $7/hr tiers
- Q2 2026·improvedCohere: New Rerank 4 Pro Large tier in Model Vault ($10/hr, $6,500/mo)
- Week 25, 2026·packagingCohere adds Coding Models tier, expanding to 4 product categories
- Week 19, 2026·improvedCohere Adds Rerank 4 Pro Large to Model Vault at $10/hr