Model comparison
Command R7B vs DeepSeek V4 Flash
DeepSeek V4 Flash has enough public results to be ranked (#32); Command R7B does not yet, so treat this comparison as directional.
Last verified . 0 shared benchmarks.
Cohere
DeepSeekSummary
- Command R7B is cheaper at $0.0375 / $0.15 per million input/output tokens, against $0.15 / $0.60 for DeepSeek V4 Flash.
- DeepSeek V4 Flash accepts more context: 1M tokens versus 128K.
Side by side
| Command R7B | DeepSeek V4 Flash | |
|---|---|---|
| Provider | Cohere | DeepSeek |
| Noometry Index | 32.1 | 53.6 |
| Released | 2024-12-02 | 2026-04-24 |
| Weights | Open | Open |
| Context window | 128K | 1M |
| Max output | 4K | 393K |
| Input $ / M tokens | $0.0375 | $0.15 |
| Output $ / M tokens | $0.15 | $0.60 |
| Results tracked | 1 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Command R7B: —, DeepSeek V4 Flash: 47.6 (#56)
| Benchmark | Command R7B | DeepSeek V4 Flash |
|---|---|---|
| FrontierCode | — | 18.8% |
| LMArena WebDev | — | 1581 |
| SciCode | — | 49.9% |
| WeirdML | — | 63% |
| LMArena Coding | — | 1457 |
| ALE-Bench | — | 1,306 |
Agentic & Tool Use Not comparable
Command R7B: 25.9, DeepSeek V4 Flash: —
| Benchmark | Command R7B | DeepSeek V4 Flash |
|---|---|---|
| Berkeley Function Calling Leaderboard | 32.1% | — |
Reasoning Not comparable
Command R7B: —, DeepSeek V4 Flash: 53.7 (#25)
| Benchmark | Command R7B | DeepSeek V4 Flash |
|---|---|---|
| ARC-AGI-2 | — | 61.4% |
| SimpleBench | — | 61.1% |
| Kagi LLM Benchmark | — | 52.2% |
| NYT Connections (extended) | — | 89.6% |
| ARC-AGI-1 | — | 89% |
| CritPt | — | 16.6% |
| Chess Puzzles | — | 33% |
| LMArena Hard Prompts | — | 1443 |
| Mystery Game Puzzles | — | 34% |
| DTBench | — | 90.9% |
| LMCA | — | 41.7% |
| Epoch Capabilities Index | — | 154.51 |
Math Not comparable
Command R7B: —, DeepSeek V4 Flash: 60.0 (#31)
| Benchmark | Command R7B | DeepSeek V4 Flash |
|---|---|---|
| FrontierMath (Tiers 1-3) | — | 57.5% |
| FrontierMath Tier 4 | — | 24.4% |
| MathArena Final-Answer Competitions | — | 76.5% |
| OTIS Mock AIME 2024-2025 | — | 94.4% |
| ProofBench | — | 56% |
| LMArena Math | — | 1427 |
Knowledge Not comparable
Command R7B: —, DeepSeek V4 Flash: 55.5 (#43)
| Benchmark | Command R7B | DeepSeek V4 Flash |
|---|---|---|
| GPQA Diamond | — | 91% |
| SimpleQA Verified | — | 33.6% |
| LMArena Expert | — | 1441 |
Multilingual Not comparable
Command R7B: —, DeepSeek V4 Flash: 53.2 (#71)
| Benchmark | Command R7B | DeepSeek V4 Flash |
|---|---|---|
| LMArena Non-English | — | 1420 |
| LMArena Chinese | — | 1467 |
| LMArena French | — | 1438 |
| LMArena German | — | 1418 |
| LMArena Japanese | — | 1403 |
| LMArena Korean | — | 1383 |
| LMArena Russian | — | 1428 |
| LMArena Spanish | — | 1437 |
Instruction Following Not comparable
Command R7B: —, DeepSeek V4 Flash: 75.1 (#76)
| Benchmark | Command R7B | DeepSeek V4 Flash |
|---|---|---|
| LMArena Instruction Following | — | 1421 |
Long Context Not comparable
Command R7B: —, DeepSeek V4 Flash: 43.8 (#81)
| Benchmark | Command R7B | DeepSeek V4 Flash |
|---|---|---|
| LMArena Longer Query | — | 1435 |
Writing & Preference Not comparable
Command R7B: —, DeepSeek V4 Flash: 64.0 (#59)
| Benchmark | Command R7B | DeepSeek V4 Flash |
|---|---|---|
| LMArena Text | — | 1432 |
| LMArena Creative Writing | — | 1403 |
| EQ-Bench Creative Writing | — | 1559 |
| LMArena Multi-Turn | — | 1449 |
Frequently asked questions
Is Command R7B better than DeepSeek V4 Flash?
DeepSeek V4 Flash has enough public results to be ranked (#32); Command R7B does not yet, so treat this comparison as directional.
Which is cheaper, Command R7B or DeepSeek V4 Flash?
Command R7B is cheaper. It lists at $0.0375 per million input tokens and $0.15 per million output tokens; DeepSeek V4 Flash lists at $0.15 and $0.60.
Which has the bigger context window?
DeepSeek V4 Flash does, with 1M tokens against 128K.
How many benchmarks do Command R7B and DeepSeek V4 Flash share?
0 benchmarks have published results for both models. Command R7B has 1 scored results on Noometry and DeepSeek V4 Flash has 41.