Noometry

Cohere, open weights

Command R+

Command R+ by Cohere ranks 257th of 354 ranked models on the Noometry Index as of October 2026, with a score of 32.4. Its strongest category is knowledge, where it ranks 169th. API pricing starts at $2.50 per million input tokens and $10 per million output tokens, with a 128K-token context window.

Last verified

Noometry rank
#257 of 354
Index score
32.4
Evidence
Confirmed 34 results
Provider
Cohere
Released
August 30, 2024
Weights
Open weights
Reasoning
No
Context window
128K
Max output
4K
Input price
$2.50 / M
Output price
$10 / M
Blended price
$4.38 / M
Output speed
Not measured
Value
#195 of 219
Knowledge cutoff
June 2024
Input
text

Category scores

Each category score combines every public result we have in that category.

  1. Coding 29.1
  2. Reasoning 9.2
  3. Math 28.9
  4. Knowledge 36.4
  5. Multilingual 38.6
  6. Instruction Following 60.0
  7. Long Context 37.3
  8. Writing & Preference 43.5
CategoryScoreRankResults
Coding29.1#3094
Reasoning9.2#3446
Math28.9#2422
Knowledge36.4#1692
Multilingual38.6#2271
Instruction Following60.0#2542
Long Context37.3#2191
Writing & Preference43.5#2284

Strengths and weaknesses

Categories where Command R+ places highest and lowest among the models ranked in each, with its score against that category's median.

Strongest categories

CategoryScorevs medianRank
Knowledge36.4−0.9#169 of 314, top 54%
Writing & Preference43.5−10.3#228 of 312, top 74%
Long Context37.3−3.6#219 of 296, top 74%

Weakest categories

CategoryScorevs medianRank
Reasoning9.2−14.4#344 of 350, top 99%
Coding29.1−9.6#309 of 340, top 91%
Instruction Following60.0−11.3#254 of 305, top 84%

Closest competitors

The models ranked just above and below Command R+. When scores are this close, price and speed are often the better way to choose.

ModelRankScoreBlended $/MSpeed
Qwen1.5-14B#25332.7——Compare
Olmo 2 0325 32b Instruct#25432.7——Compare
gpt-oss-20b#25532.5$0.03696Compare
Laguna M.1#25632.5——Compare
Granite 3.1 8b Instruct#25832.4——Compare
Pixtral Large#25932.2$3—Compare
Falcon-180B#26032.2——Compare
Gemini 1.5 Pro (May 2024)#26132.1——Compare

Sponsored placements are available on pages like this one. Advertise on Noometry

Benchmark results

Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.

Coding

BenchmarkScorePositionSettingSourceDate
BigCodeBench Instruct33.8%#49 of 64, top 77%BigCodeBench2024-04-04
LiveBench Coding19.1%#36 of 39, top 93%Epoch AI
LMArena Coding1187#245 of 294, top 84%LMArena2026-10-08
LMArena Coding1172LMArena2026-10-08
BigCodeBench Complete41.9%#49 of 66, top 75%BigCodeBench2024-04-04
HumanEval+56.7%#31 of 45, top 69%EvalPlus
MBPP+63.5%#21 of 38, top 56%EvalPlus

Reasoning

BenchmarkScorePositionSettingSourceDate
SimpleBench17.4%#76 of 77, top 99%Epoch AI
LiveBench Reasoning24.8%#35 of 39, top 90%Epoch AI
LMArena Hard Prompts1186#244 of 297, top 83%LMArena2026-10-08
LMArena Hard Prompts1166LMArena2026-10-08
DTBench54.9%#123 of 151, top 82%Epoch AI
LiveBench Data Analysis38.1%#31 of 39, top 80%Epoch AI
LMCA5%#123 of 125, top 99%Epoch AI
Epoch Capabilities Index119.34#171 of 213, top 81%Epoch AI2024-08-30
LiveBench31.8%#33 of 39, top 85%Epoch AI

Math

BenchmarkScorePositionSettingSourceDate
LiveBench Math21.3%#34 of 39, top 88%Epoch AI
LMArena Math1164LMArena2026-10-08
LMArena Math1188#236 of 285, top 83%LMArena2026-10-08

Knowledge

BenchmarkScorePositionSettingSourceDate
Vectara Hallucination Rate (lower is better)6.9%#26 of 96, top 28%Vectara Hallucination Leaderboard
LMArena Expert1174#227 of 273, top 84%LMArena2026-10-08
LMArena Expert1153LMArena2026-10-08
MMLU69.4%#47 of 81, top 59%Epoch AI

Multilingual

BenchmarkScorePositionSettingSourceDate
LMArena Non-English1190LMArena2026-10-08
LMArena Non-English1216#227 of 297, top 77%LMArena2026-10-08
LMArena Chinese1226#220 of 285, top 78%LMArena2026-10-08
LMArena Chinese1191LMArena2026-10-08
LMArena French1209#191 of 223, top 86%LMArena2026-10-08
LMArena German1195LMArena2026-10-08
LMArena German1216#187 of 231, top 81%LMArena2026-10-08
LMArena Japanese1166#170 of 211, top 81%LMArena2026-10-08
LMArena Japanese1155LMArena2026-10-08
LMArena Korean1138#184 of 213, top 87%LMArena2026-10-08
LMArena Russian1227#223 of 283, top 79%LMArena2026-10-08
LMArena Russian1206LMArena2026-10-08
LMArena Spanish1189#197 of 226, top 88%LMArena2026-10-08

Instruction Following

BenchmarkScorePositionSettingSourceDate
LiveBench Instruction Following57.6%#31 of 39, top 80%Epoch AI
LMArena Instruction Following1197#238 of 298, top 80%LMArena2026-10-08
LMArena Instruction Following1180LMArena2026-10-08

Long Context

BenchmarkScorePositionSettingSourceDate
LMArena Longer Query1230#230 of 291, top 80%LMArena2026-10-08
LMArena Longer Query1202LMArena2026-10-08

Writing & Preference

BenchmarkScorePositionSettingSourceDate
LMArena Text1204LMArena2026-10-08
LMArena Text1229#231 of 297, top 78%LMArena2026-10-08
LMArena Creative Writing1235#218 of 295, top 74%LMArena2026-10-08
LMArena Creative Writing1201LMArena2026-10-08
LMArena Multi-Turn1192LMArena2026-10-08
LMArena Multi-Turn1213#235 of 295, top 80%LMArena2026-10-08
LiveBench Language29.7%#27 of 39, top 70%Epoch AI

API pricing by provider

RouteInput $/MOutput $/MCached input $/MChecked
cohere$2.50$10—2026-10-10
openrouter$2.50$10—2026-10-10

All Cohere API prices → Estimate your cost →

Compare Command R+

Other Cohere models

Frequently asked questions

How good is Command R+?

Command R+ by Cohere ranks 257th of 354 ranked models on the Noometry Index as of October 2026, with a score of 32.4. Its strongest category is knowledge, where it ranks 169th. API pricing starts at $2.50 per million input tokens and $10 per million output tokens, with a 128K-token context window.

How much does Command R+ cost?

Command R+ costs $2.50 per million input tokens and $10 per million output tokens on Cohere's own API.

What is Command R+'s context window?

Command R+ accepts up to 128K tokens of input and can write up to 4K tokens in one response.

Is Command R+ open source?

Yes. Command R+'s weights are downloadable; check the license for commercial terms.

What are Command R+'s strengths and weaknesses?

Relative to other ranked models, Command R+ places best in knowledge, writing & preference, long context and lowest in reasoning, coding, instruction following.

What is Command R+ best at?

Its best category is knowledge, where it ranks 169th on Noometry.