Command R+ Base: pricing, context window and benchmarks
Command R+ Base is a open-weight model from Cohere. It offers a 65,536-token context window, costs $3.24 per million input tokens and $14.70 per million output tokens, and averages 71.6 across the index's benchmark suite.
Pricing
Command R+ Base costs $3.24 per million input tokens and $14.70 per million output tokens. A blended 3:1 input-to-output mix works out to about $6.11 per million tokens. Cached input is listed at $0.41 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $3.24 | per 1M tokens |
| Output | $14.70 | per 1M tokens |
| Cached input | $0.41 | per 1M tokens |
| Blended 3:1 | $6.11 | estimate per 1M tokens |
| 1M in + 1M out | $17.94 | estimate |
Benchmarks
Scores below are modelled estimates on a 0–100 scale. Use them to compare relative position, not as official vendor results.
| Benchmark | Score | Rank | Percentile | Relative |
|---|---|---|---|---|
| MMLU-Pro | 76.1 | 301 | 50th | |
| GPQA Diamond | 66.8 | 334 | 44th | |
| SWE-bench Verified | 60.7 | 310 | 48th | |
| HumanEval | 76.6 | 314 | 48th | |
| MATH-500 | 74.9 | 325 | 46th | |
| AIME 2025 | 63.5 | 362 | 40th | |
| LiveCodeBench | 62.4 | 344 | 43th | |
| IFEval | 79.2 | 349 | 42th | |
| TruthfulQA | 73.9 | 356 | 41th | |
| HellaSwag | 83.7 | 289 | 52th | |
| MMMU | 69.5 | 315 | 48th | |
| Arena Elo | 1375.6 | 290 | 52th |
What Command R+ Base is good for
Command R+ Base is an open-weight model, so it can be self-hosted or run through a third-party inference provider. That matters for data residency, fine-tuning and long-run cost control. At $3.24 per million input tokens (hosted list price) and a 65,536-token context window, it is a practical option for teams that want more control than a closed API allows.
Strengths
- Input price of $3.24 per million tokens is at or below the index median.
- Open weights make self-hosting, fine-tuning and data-residency review possible.
- Accepts more than text, including image input where the provider supports it.
Trade-offs
- Average benchmark score is below the index median; test it on your own task.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| DBRX Chat | Databricks | 65536 | $3.07 | 71.7 |
| OLMo 2 Instruct | Allen Institute for AI | 32768 | $3.03 | 71.7 |
| Qwen3-Max Nano | Alibaba Cloud | 65536 | $8.31 | 71.7 |
| Claude Haiku 4.5 Mini | Anthropic | 16384 | $8.76 | 71.6 |
| Command R+ Base | Cohere | 65536 | $3.24 | 71.6 |
| Auto Reasoning | OpenRouter | 65536 | $8.39 | 71.5 |
FAQ
How much does Command R+ Base cost?
Command R+ Base is listed at $3.24 per million input tokens and $14.70 per million output tokens. Cached input is $0.41 per million tokens. A blended 3:1 input-to-output mix is about $6.11 per million tokens.
What is the context window of Command R+ Base?
Command R+ Base has a context window of 65,536 tokens and can emit up to 8,192 output tokens per request in this index.
Is Command R+ Base open weights?
Yes. Command R+ Base is listed as an open-weight model under a open weights licence.
How does Command R+ Base score on benchmarks?
Command R+ Base averages 71.6 across the scored benchmarks in this index, ranking #338 of 600 models. Its strongest relative result is Arena Elo at 1375.6.
What is Command R+ Base best for?
Command R+ Base is an open-weight model, so it can be self-hosted or run through a third-party inference provider. That matters for data residency, fine-tuning and long-run cost control. At $3.24 per million input tokens (hosted list price) and a 65,536-token context window, it is a practical option for teams that want more control than a closed API allows.