Model comparison
Command R vs o1
o1 is the stronger model overall, scoring 40.9 to 31.4 on the Noometry Index. Command R costs 100× less per token, which makes it the better buy when o1's lead doesn't matter for your workload.
Last verified . 26 shared benchmarks.
CohereSummary
- They share 26 benchmarks with published results for both. Command R scores higher in 0 categories and o1 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where o1 leads 55.6 to 38.2.
- The biggest single-benchmark swing is LiveBench Reasoning: 21.9% for Command R and 91.6% for o1.
- Command R is cheaper at $0.15 / $0.60 per million input/output tokens, against $15 / $60 for o1.
- o1 accepts more context: 200K tokens versus 128K.
- Command R has downloadable open weights; the other is API-only.
Side by side
| Command R | o1 | |
|---|---|---|
| Provider | Cohere | OpenAI |
| Noometry Index | 31.4 | 40.9 |
| Released | 2024-08-30 | 2024-09-12 |
| Weights | Open | Proprietary |
| Context window | 128K | 200K |
| Max output | 4K | 100K |
| Input $ / M tokens | $0.15 | $15 |
| Output $ / M tokens | $0.60 | $60 |
| Results tracked | 29 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o1 leads
Command R: 29.3 (#306), o1: 46.1 (#70)
| Benchmark | Command R | o1 |
|---|---|---|
| LiveBench Coding | 17.9% | 69.7% |
| LMArena Coding | 1169 | 1367 |
| Aider Polyglot | — | 61.7% |
| WeirdML | — | 47.6% |
| BigCodeBench Instruct | 37.1% | — |
| BigCodeBench Complete | 45.2% | — |
| CadEval | — | 56% |
| HumanEval+ | — | 89% |
| MBPP+ | — | 80.2% |
Agentic & Tool Use Not comparable
Command R: —, o1: 24.6 (#117)
| Benchmark | Command R | o1 |
|---|---|---|
| Cybench | — | 10% |
| METR Time Horizons | — | 51.1% |
Reasoning o1 leads
Command R: 13.8 (#331), o1: 27.9 (#111)
| Benchmark | Command R | o1 |
|---|---|---|
| LiveBench Reasoning | 21.9% | 91.6% |
| LMArena Hard Prompts | 1164 | 1371 |
| DTBench | 46.4% | 74.7% |
| LiveBench Data Analysis | 33.3% | 65.5% |
| LMCA | 9.2% | 22.3% |
| LiveBench | 27.5% | 75.7% |
| SimpleBench | — | 41.7% |
| ARC-AGI-1 | — | 30.7% |
| Chess Puzzles | — | 15% |
| EnigmaEval | — | 5.7% |
| Epoch Capabilities Index | — | 141.91 |
Math o1 leads
Command R: 28.0 (#246), o1: 36.1 (#175)
| Benchmark | Command R | o1 |
|---|---|---|
| LiveBench Math | 19.4% | 80.3% |
| LMArena Math | 1155 | 1388 |
| FrontierMath (Tiers 1-3) | — | 14.7% |
| OTIS Mock AIME 2024-2025 | — | 73.3% |
| MATH Level 5 | — | 94.7% |
| FrontierMath (Feb 2025 set) | — | 9.3% |
Knowledge o1 leads
Command R: 31.0 (#221), o1: 41.5 (#110)
| Benchmark | Command R | o1 |
|---|---|---|
| LMArena Expert | 1138 | 1361 |
| GPQA Diamond | — | 76.8% |
| Humanity's Last Exam | — | 8% |
| SimpleQA Verified | — | 41.1% |
| Confabulations | — | 11.7% |
| MMLU | 65.2% | — |
Multimodal Not comparable
Command R: —, o1: 34.2 (#93)
| Benchmark | Command R | o1 |
|---|---|---|
| LMArena Vision | — | 1168 |
| GeoBench | — | 80% |
| VPCT | — | 37% |
| SpatialViz-Bench | — | 41.4% |
Multilingual o1 leads
Command R: 35.7 (#245), o1: 48.6 (#142)
| Benchmark | Command R | o1 |
|---|---|---|
| LMArena Non-English | 1174 | 1358 |
| LMArena Chinese | 1182 | 1394 |
| LMArena French | 1162 | 1344 |
| LMArena German | 1176 | 1337 |
| LMArena Japanese | 1143 | 1346 |
| LMArena Korean | 1163 | 1396 |
| LMArena Russian | 1174 | 1356 |
| LMArena Spanish | 1151 | 1345 |
Instruction Following o1 leads
Command R: 58.1 (#261), o1: 74.8 (#86)
| Benchmark | Command R | o1 |
|---|---|---|
| LiveBench Instruction Following | 55.6% | 81.5% |
| LMArena Instruction Following | 1167 | 1367 |
Long Context o1 leads
Command R: 36.3 (#231), o1: 50.3 (#9)
| Benchmark | Command R | o1 |
|---|---|---|
| LMArena Longer Query | 1198 | 1378 |
| Fiction.LiveBench | — | 83.3% |
Writing & Preference o1 leads
Command R: 38.2 (#254), o1: 55.6 (#144)
| Benchmark | Command R | o1 |
|---|---|---|
| LMArena Text | 1187 | 1366 |
| LMArena Creative Writing | 1170 | 1348 |
| LMArena Multi-Turn | 1163 | 1369 |
| LiveBench Language | 16.7% | 65.4% |
| Short-Story Creative Writing | — | 70.2% |
Frequently asked questions
Is Command R better than o1?
o1 is the stronger model overall, scoring 40.9 to 31.4 on the Noometry Index. Command R costs 100× less per token, which makes it the better buy when o1's lead doesn't matter for your workload.
Which is cheaper, Command R or o1?
Command R is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; o1 lists at $15 and $60.
Is Command R or o1 better for coding?
o1 scores higher on coding benchmarks: 46.1 versus 29.3 in the Noometry coding category.
Which has the bigger context window?
o1 does, with 200K tokens against 128K.
How many benchmarks do Command R and o1 share?
26 benchmarks have published results for both models. Command R has 29 scored results on Noometry and o1 has 52.