Deep Cogito8BQ4_K_M 下约 6 GB 显存

Cogito v1 8B — 显存、速度与本地部署

作者: Jakub Rusinowski · 最后更新:

Mid-tier Cogito variant optimized for 8 GB VRAM GPUs. Dynamic reasoning gives it reasoning quality comparable to Llama 3.1 70B on math benchmarks, while running on a single RTX 4060. Best price-to-performance reasoning model at the 8B tier.

Cogito v1 8B 在 Q4_K_M 下约需 6 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78
创意70
编程74

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
4.5 GB~155 tok/s可运行
Q3_K_M
3.41 bpw
5.3 GB~134 tok/s可运行
Q4_K_M
4.83 bpw
6.7 GB~107 tok/s可运行
Q5_K_M
5.67 bpw
7.5 GB~96 tok/s可运行
Q6_K
6.56 bpw
8.4 GB~86 tok/s可运行
Q8_0
8.50 bpw
10.4 GB~70 tok/s可运行
F16
16.00 bpw
17.9 GB~41 tok/s可运行

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. Cogito v1 8B 显存计算器 →

运行 Cogito v1 8B

目录中能运行 Cogito v1 8B 的最便宜 GPU 是 Intel Arc B570 (10 GB).

购买此硬件 Intel Arc B570 10GB — 10 GB VRAM · 150 W board power立即云端部署 RunPod 上的 RTX 4090 — 低至 $0.34/小时 · 价格核实于 2026-07

或在 Vast.ai 比较,低至 $0.35/小时 (typical low · varies)

作为亚马逊联盟成员,我们从符合条件的购买中获得收入。云 GPU 链接为推荐链接——我们可能获得佣金,您无需额外付费。

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。
Intel Arc B570 10GB
10 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 Cogito v1 8B

安装 Ollama,然后运行:

ollama run cogito:8b
Hugging Face 上的权重: deepcogito/Cogito-v1-8B-Instruct ↗

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
8 Billion
上下文窗口
32,000
架构
Hybrid Reasoning Transformer
提供商
Deep Cogito
许可证
Apache 2.0
Practical deployment
Laptop
规格量化
Q4_K_M
系统内存
16 GB
记录更新于
2026-03-20
许可证Apache-2.0允许商业使用

Commercial use permitted. No usage restrictions beyond attribution.

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reasoningmathlocal inferenceconsumer gpu
基准分数来源
MMLU75.3 / 100 %已发布
MATH65.8 / 100 %已发布
HumanEval70.2 / 100 %已发布

我的 GPU 能运行 Cogito v1 8B 吗?

Cogito v1 的其他尺寸

Cogito v1 8B — 常见问题

How much VRAM does Cogito v1 8B need?

About 6 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 Cogito v1 8B run on an RTX 4090 (24 GB)?

Yes. Cogito v1 8B needs about 6 GB at Q4_K_M, inside a 24 GB card, at an estimated 107 tokens/sec.

How do I run Cogito v1 8B locally?

Install Ollama and run `ollama run cogito:8b`. That pulls the weights and starts a local OpenAI-compatible endpoint; after the download nothing leaves the machine.

What other sizes does Cogito v1 come in?

Cogito v1 3B (3 GB), Cogito v1 8B (6 GB), Cogito v1 14B (9 GB), Cogito v1 32B (20 GB), Cogito v1 70B (43 GB). Every size shares the family's training and licence; the larger ones score higher and need proportionally more memory.