Model comparison
Command R vs Qwen3.8 Max
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 31.4 on the Noometry Index. Command R costs 11× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Last verified . 19 shared benchmarks.
Cohere
Alibaba (Qwen)Summary
- They share 19 benchmarks with published results for both. Command R scores higher in 0 categories and Qwen3.8 Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.8 Max leads 73.2 to 28.0.
- The biggest single-benchmark swing is DTBench: 46.4% for Command R and 92% for Qwen3.8 Max.
- Command R is cheaper at $0.15 / $0.60 per million input/output tokens, against $2 / $6 for Qwen3.8 Max.
- Qwen3.8 Max accepts more context: 1M tokens versus 128K.
- Command R has downloadable open weights; the other is API-only.
Side by side
| Command R | Qwen3.8 Max | |
|---|---|---|
| Provider | Cohere | Alibaba (Qwen) |
| Noometry Index | 31.4 | 56.8 |
| Released | 2024-08-30 | 2026-08-02 |
| Weights | Open | Proprietary |
| Context window | 128K | 1M |
| Max output | 4K | 131K |
| Input $ / M tokens | $0.15 | $2 |
| Output $ / M tokens | $0.60 | $6 |
| Results tracked | 29 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 Max leads
Command R: 29.3 (#306), Qwen3.8 Max: 53.5 (#29)
| Benchmark | Command R | Qwen3.8 Max |
|---|---|---|
| LMArena Coding | 1169 | 1502 |
| DeepSWE | — | 57.5% |
| LMArena WebDev | — | 1674 |
| FrontierSWE | — | 17.8% |
| SciCode | — | 53.2% |
| BigCodeBench Instruct | 37.1% | — |
| LiveBench Coding | 17.9% | — |
| BigCodeBench Complete | 45.2% | — |
Agentic & Tool Use Not comparable
Command R: —, Qwen3.8 Max: 45.4 (#14)
| Benchmark | Command R | Qwen3.8 Max |
|---|---|---|
| APEX-Agents | — | 63.3% |
| τ²-bench Banking | — | 55.1% |
| GDP.pdf | — | 23.2% |
Reasoning Qwen3.8 Max leads
Command R: 13.8 (#331), Qwen3.8 Max: 54.4 (#26)
| Benchmark | Command R | Qwen3.8 Max |
|---|---|---|
| LMArena Hard Prompts | 1164 | 1496 |
| DTBench | 46.4% | 92% |
| LMCA | 9.2% | 46.2% |
| NYT Connections (extended) | — | 88.3% |
| CritPt | — | 20% |
| Chess Puzzles | — | 40% |
| LiveBench Reasoning | 21.9% | — |
| Mystery Game Puzzles | — | 38% |
| LiveBench Data Analysis | 33.3% | — |
| Epoch Capabilities Index | — | 156.41 |
| LiveBench | 27.5% | — |
Math Qwen3.8 Max leads
Command R: 28.0 (#246), Qwen3.8 Max: 73.2 (#20)
| Benchmark | Command R | Qwen3.8 Max |
|---|---|---|
| LMArena Math | 1155 | 1499 |
| FrontierMath (Tiers 1-3) | — | 74.7% |
| FrontierMath Tier 4 | — | 46.3% |
| OTIS Mock AIME 2024-2025 | — | 100% |
| ProofBench | — | 58% |
| LiveBench Math | 19.4% | — |
Knowledge Qwen3.8 Max leads
Command R: 31.0 (#221), Qwen3.8 Max: 61.7 (#27)
| Benchmark | Command R | Qwen3.8 Max |
|---|---|---|
| LMArena Expert | 1138 | 1507 |
| GPQA Diamond | — | 92.7% |
| SimpleQA Verified | — | 47.3% |
| MMLU | 65.2% | — |
Multimodal Not comparable
Command R: —, Qwen3.8 Max: 37.2 (#75)
| Benchmark | Command R | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | — | 1314 |
| Furniture Assembly | — | 20% |
Multilingual Qwen3.8 Max leads
Command R: 35.7 (#245), Qwen3.8 Max: 56.7 (#18)
| Benchmark | Command R | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1174 | 1472 |
| LMArena Chinese | 1182 | 1538 |
| LMArena French | 1162 | 1503 |
| LMArena German | 1176 | 1483 |
| LMArena Japanese | 1143 | 1467 |
| LMArena Korean | 1163 | 1461 |
| LMArena Russian | 1174 | 1481 |
| LMArena Spanish | 1151 | 1492 |
Instruction Following Qwen3.8 Max leads
Command R: 58.1 (#261), Qwen3.8 Max: 77.6 (#17)
| Benchmark | Command R | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1167 | 1479 |
| LiveBench Instruction Following | 55.6% | — |
Long Context Qwen3.8 Max leads
Command R: 36.3 (#231), Qwen3.8 Max: 45.6 (#31)
| Benchmark | Command R | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1198 | 1489 |
Writing & Preference Qwen3.8 Max leads
Command R: 38.2 (#254), Qwen3.8 Max: 67.1 (#30)
| Benchmark | Command R | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1187 | 1483 |
| LMArena Creative Writing | 1170 | 1479 |
| LMArena Multi-Turn | 1163 | 1489 |
| LiveBench Language | 16.7% | — |
Frequently asked questions
Is Command R better than Qwen3.8 Max?
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 31.4 on the Noometry Index. Command R costs 11× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Which is cheaper, Command R or Qwen3.8 Max?
Command R is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Qwen3.8 Max lists at $2 and $6.
Is Command R or Qwen3.8 Max better for coding?
Qwen3.8 Max scores higher on coding benchmarks: 53.5 versus 29.3 in the Noometry coding category.
Which has the bigger context window?
Qwen3.8 Max does, with 1M tokens against 128K.
How many benchmarks do Command R and Qwen3.8 Max share?
19 benchmarks have published results for both models. Command R has 29 scored results on Noometry and Qwen3.8 Max has 39.