vs
0:1
overall capability wins (1 ties across 2 scored). Command R+ (08-2024) leads overall.
- Pricing3:0
- Speed2:0
- Specs0:1
Verdict
Command R+ (08-2024) offers ~31.3x larger context (128K vs 4K); Command Light is ~11.7x cheaper on a blended token basis than Command R+ (08-2024).
- Command R+ (08-2024) offers ~31.3x larger context (128K vs 4K)
- Command Light is ~11.7x cheaper on a blended token basis than Command R+ (08-2024)
Cost for 1M in + 250K out
Illustrative chat workload at primary-provider list prices.
- Command Light
- $0.45
- Command R+ (08-2024)
- $5
Command Light is about 11.1x cheaper on this workload.
Price per 1M tokens
Full comparison
Organization
Cohere
Cohere
Family
Command
Command
License
Proprietary
Proprietary
Open weights
No
No
Release
Nov 2, 2023
Aug 1, 2024
Knowledge cutoff
-
2024-06-01
API / provider
Cohere
Cohere
Modalities
text → text
text → text
Specs
Context window
4K
128K
Max output
4K
4K
Parameters
—
—
Pricing
Input $/1M
$0.30
$2.50
Output $/1M
$0.60
$10
Blended $/1M (3∶1)
$0.38
$4.38
Speed
tok/s
120
75
TTFT (s)
0.2
0.4
Reasoning
MMLU-Pro
—
—
GPQA Diamond
—
—
Humanity's Last Exam
—
—
AIME 2025
—
—
MATH-500
—
—
Humanity's Last Exam (with tools)
—
—
AA-Omniscience Accuracy
—
—
AA-LCR v1.1
—
—
CritPt
—
—
MMMU-Pro
—
—
IFBench
—
—
Chartography
—
—
Chartography (With Tools)
—
—
Coding
SWE-bench Verified
—
—
SWE-bench Pro
—
—
SWE-bench Multilingual
—
—
LiveCodeBench
—
—
Terminal-Bench 2.1
—
—
Aider Polyglot
—
—
Terminal-Bench 3
—
—
BigCodeBench
—
—
SciCode
—
—
CursorBench
—
—
SWE-Rebench
—
—
NL2Repo-Bench
—
—
DeepSWE
—
—
WebDev Arena
—
—
Terminal-Bench 4.0
—
—
CursorBench 4.0
—
—
FrontierCode v1.1 (Main)
—
—
Arena
LMArena Elo
—
1,276
| Metric | Command Light | Command R+ (08-2024) | Delta |
|---|---|---|---|
| Identity | |||
| Organization | Cohere | Cohere | — |
| Family | Command | Command | — |
| License | Proprietary | Proprietary | — |
| Open weights | No | No | — |
| Release | Nov 2, 2023 | Aug 1, 2024 | — |
| Knowledge cutoff | - | 2024-06-01 | — |
| API / provider | Cohere | Cohere | — |
| Modalities | text → text | text → text | — |
| Specs | |||
| Context window | 4K | 128K | B +124K |
| Max output | 4K | 4K | tie |
| Parameters | — | — | — |
| Pricing | |||
| Input $/1M | $0.30 | $2.50 | A +$2.20 |
| Output $/1M | $0.60 | $10 | A +$9.40 |
| Blended $/1M (3∶1) | $0.38 | $4.38 | A +$4 |
| Speed | |||
| tok/s | 120 | 75 | A +45 |
| TTFT (s) | 0.2 | 0.4 | A +0.20 |
| Reasoning | |||
| MMLU-Pro | — | — | — |
| GPQA Diamond | — | — | — |
| Humanity's Last Exam | — | — | — |
| AIME 2025 | — | — | — |
| MATH-500 | — | — | — |
| Humanity's Last Exam (with tools) | — | — | — |
| AA-Omniscience Accuracy | — | — | — |
| AA-LCR v1.1 | — | — | — |
| CritPt | — | — | — |
| MMMU-Pro | — | — | — |
| IFBench | — | — | — |
| Chartography | — | — | — |
| Chartography (With Tools) | — | — | — |
| Coding | |||
| SWE-bench Verified | — | — | — |
| SWE-bench Pro | — | — | — |
| SWE-bench Multilingual | — | — | — |
| LiveCodeBench | — | — | — |
| Terminal-Bench 2.1 | — | — | — |
| Aider Polyglot | — | — | — |
| Terminal-Bench 3 | — | — | — |
| BigCodeBench | — | — | — |
| SciCode | — | — | — |
| CursorBench | — | — | — |
| SWE-Rebench | — | — | — |
| NL2Repo-Bench | — | — | — |
| DeepSWE | — | — | — |
| WebDev Arena | — | — | — |
| Terminal-Bench 4.0 | — | — | — |
| CursorBench 4.0 | — | — | — |
| FrontierCode v1.1 (Main) | — | — | — |
| Arena | |||
| LMArena Elo | — | 1,276 | — |
Command Light
- Context
- 4K / 4K out
- Parameters
- —
- Price
- $0.30 / $0.60
- Speed
- 120 tok/s · 0.2s TTFT
- Modalities
- text → text
- License
- Proprietary
Cohere's budget Command SKU for classification and short RAG answers.
Command R+ (08-2024)
- Context
- 128K / 4K out
- Parameters
- —
- Price
- $2.50 / $10
- Speed
- 75 tok/s · 0.4s TTFT
- Modalities
- text → text
- License
- Proprietary
Cohere's RAG/agent flagship before Command A — strong multi-hop tool use at 128K context; still widely deployed on Bedrock and Azure.
Benchmark charts
Winner bars are emphasized. Per-benchmark deltas sit above each chart.
These two models have no overlapping published benchmarks in our dataset. Compare specs and pricing instead.