Cohere104BQ4_K_M 下约 64 GB 显存

Command R+ (104B) — 显存、速度与本地部署

作者: Jakub Rusinowski · 最后更新:

Cohere's flagship model for complex enterprise RAG. Handles multi-hop retrieval, grounded citations, and complex tool orchestration. Outperforms GPT-4 Turbo on RAG benchmarks.

Command R+ (104B) 在 Q4_K_M 下约需 64 GB 显存——量化权重加框架开销,不含 KV 缓存。在 Apple Silicon 上,这部分来自统一内存。

逻辑94
创意87
编程88

按量化级别的显存与速度

计算基准:NVIDIA RTX 4090 (24 GB)。包含 8K 上下文的 KV 缓存,因此数值高于上方的主数字。

量化显存速度(估算)适配
Q2_K
2.63 bpw
37.1 GB~3 tok/s需卸载
Q3_K_M
3.41 bpw
47.3 GB~2 tok/s需卸载
Q4_K_M
4.83 bpw
65.7 GB—放不下
Q5_K_M
5.67 bpw
76.7 GB—放不下
Q6_K
6.56 bpw
88.2 GB—放不下
Q8_0
8.50 bpw
113.4 GB—放不下
F16
16.00 bpw
210.9 GB—放不下

黑色标记 = NVIDIA RTX 4090 (24 GB) 上的可用显存。 估算来自内存带宽屋顶线模型,详见 方法说明页. Command R+ (104B) 显存计算器 →

运行 Command R+ (104B)

目录中能运行 Command R+ (104B) 的最便宜 GPU 是 NVIDIA DGX Spark (128 GB).

联盟营销声明: 本页部分链接为联盟推广链接——如果你通过它们购买,LLM Configurator 可能会获得佣金,而你无需支付任何额外费用。作为亚马逊联盟成员(Amazon Associate),LLM Configurator 会从符合条件的购买中获得收益。
NVIDIA DGX Spark (128GB)
128 GB VRAM · 150 W board power
2026年价格波动较大——请以当前商品页价格为准。

如何运行 Command R+ (104B)

安装 Ollama,然后运行:

ollama run command-r-plus
Hugging Face 上的权重: CohereForAI/c4ai-command-r-plus ↗

规格

Curated — A hand-written entry from before this catalogue recorded its sources. The figures are long-standing but their provenance is not on file.

参数量
104 Billion
上下文窗口
128,000
架构
Dense
提供商
Cohere
许可证
CC-BY-NC
Practical deployment
Workstation (large unified memory or pro GPU)
规格量化
Q4_K_M
系统内存
128 GB
记录更新于
2024-04-04
许可证CC-BY-NC-4.0仅限研究/非商业用途

Research / non-commercial only — this licence does NOT permit shipping a commercial product.

质量与使用场景

评分由模型作者或独立评测方发布——衡量质量而非吞吐量,并非我们实测。

最适合enterprise ragagentcomplex retrieval

我的 GPU 能运行 Command R+ (104B) 吗?

Command R Family 的其他尺寸

Command R+ (104B) — 常见问题

How much VRAM does Command R+ (104B) need?

About 64 GB at Q4_K_M — quantized weights plus framework overhead, before any KV cache. The cache grows with context length and is added on top; the table above folds it in. Apple Silicon counts unified memory toward the same figure.

Does Command R+ (104B) run on an RTX 4090 (24 GB)?

No. Command R+ (104B) needs about 64 GB at Q4_K_M, more than a single RTX 4090's 24 GB. It needs a larger card, several GPUs, or Apple Silicon with enough unified memory — or it runs with part of the weights offloaded to system RAM, which is much slower.

How do I run Command R+ (104B) locally?

Install Ollama and run `ollama run command-r-plus`. That pulls the weights and starts a local OpenAI-compatible endpoint; after the download nothing leaves the machine.

What other sizes does Command R Family come in?

Command R (35B) (22 GB), Command R+ (104B) (64 GB). Every size shares the family's training and licence; the larger ones score higher and need proportionally more memory.