K2 Horizon 375B-A23B — 显存、速度与本地部署
作者： Jakub Rusinowski · 最后更新：
The flagship of the K2 Horizon family: a sparse MoE that stores 375B parameters and runs 23B per token (192 experts, 8 routed), with a native 524,288-token context. Apache-2.0, released with its data, training code and intermediate checkpoints. The checkpoint holds 379.2B tensors, which is what must be resident. The model card's serving routes are vLLM and SGLang, validated on 8× H200.
K2 Horizon 375B-A23B 在 Q4_K_M 下约需 230 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。
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 | 显存 | 速度（估算） | 适配 |
|---|---|---|---|---|
| Q2_K 2.63 bpw | 127.5 GB | — | 放不下 | |
| Q3_K_M 3.41 bpw | 164.5 GB | — | 放不下 | |
| Q4_K_M 4.83 bpw | 231.8 GB | — | 放不下 | |
| Q5_K_M 5.67 bpw | 271.6 GB | — | 放不下 | |
| Q6_K 6.56 bpw | 313.8 GB | — | 放不下 | |
| Q8_0 8.50 bpw | 405.7 GB | — | 放不下 | |
| F16 16.00 bpw | 761.2 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. K2 Horizon 375B-A23B 显存计算器 →
运行 K2 Horizon 375B-A23B
目录中能运行 K2 Horizon 375B-A23B 的最便宜 GPU 是 Apple M3 Ultra (512 GB).
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。
如何运行 K2 Horizon 375B-A23B
No Ollama library tag and no published GGUF for this size. The model card's serving routes are vLLM and SGLang.
规格
Verified — Checked against the primary source — the model card or the vendor spec page — and corroborated by a second independent source.
- 参数量
- 375 Billion (23B active)
- 上下文窗口
- 524,288
- 架构
- Sparse MoE (192 experts, 8 active), GQA
- 提供商
- MBZUAI Institute of Foundation Models
- 许可证
- Apache 2.0
- Availability
- Weights downloadable
- Practical deployment
- Workstation (large unified memory or pro GPU)
- 规格量化
- Q4_K_M
- 系统内存
- 256 GB
- 记录更新于
- 2026-10-13
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。
Independent evaluation — not vendor-claimed and not measured by us
| 基准 | 分数 | 来源 |
|---|---|---|
| AA Intelligence Index v4.3.2 | 31 | independent · Artificial Analysis · snapshot 2026-10-13 |
K2 Horizon 的其他尺寸
K2 Horizon 375B-A23B — 常见问题
How much VRAM does K2 Horizon 375B-A23B need?
About 230 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 K2 Horizon 375B-A23B run on an RTX 4090 (24 GB)?
No. K2 Horizon 375B-A23B needs about 230 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 K2 Horizon 375B-A23B locally?
No Ollama library tag and no published GGUF for this size. The model card's serving routes are vLLM and SGLang. Running the published tag would send your prompts to a hosted GPU rather than your own machine.
What other sizes does K2 Horizon come in?
K2 Horizon 3.7B (4 GB), K2 Horizon 7B (6 GB), K2 Horizon MoVA 36B-A4B (23 GB), K2 Horizon 375B-A23B (230 GB). Every size shares the family's training and licence; the larger ones score higher and need proportionally more memory.