Laguna XS 2.1 33B-A3B — 显存、速度与本地部署
作者: Jakub Rusinowski · 最后更新:
33B total parameters with only 3B activated per token, which is what lets a model scoring 70.9% on SWE-Bench Verified sit on a single 24GB card. Mixed sliding-window + global attention (3:1 across 40 layers), sigmoid per-head gating, FP8 KV cache, 256K context. Released under OpenMDW-1.1 — the Linux Foundation / NVIDIA permissive model-weights framework — not Apache 2.0, so check the terms if you are redistributing.
Laguna XS 2.1 33B-A3B 在 Q4_K_M 下约需 21 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)。仅含权重与开销:该模型架构未公开,因此未计入 KV 缓存。
| 量化 | 显存 | 显存 | 速度(估算) | 适配 |
|---|---|---|---|---|
| Q2_K 2.63 bpw | 11.6 GB | ~210 tok/s | 可运行 | |
| Q3_K_M 3.41 bpw | 14.9 GB | ~194 tok/s | 可运行 | |
| Q4_K_M 4.83 bpw | 20.7 GB | ~171 tok/s | 可运行 | |
| Q5_K_M 5.67 bpw | 24.2 GB | ~24 tok/s | 需卸载 | |
| Q6_K 6.56 bpw | 27.9 GB | ~22 tok/s | 需卸载 | |
| Q8_0 8.50 bpw | 35.9 GB | ~20 tok/s | 需卸载 | |
| F16 16.00 bpw | 66.8 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. Laguna XS 2.1 33B-A3B 显存计算器 →
运行 Laguna XS 2.1 33B-A3B
目录中能运行 Laguna XS 2.1 33B-A3B 的最便宜 GPU 是 AMD Radeon RX 7900 XTX (24 GB).
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。
如何运行 Laguna XS 2.1 33B-A3B
安装 Ollama,然后运行:
ollama run poolside-laguna规格
Preview — The model is released, but these specs are thin or rest on a single source. Individual fields may be wrong.
Preview — The model is released, but these specs are thin or rest on a single source. Individual fields may be wrong.
- 参数量
- 33 Billion (3B active)
- 上下文窗口
- 262,144
- 架构
- Mixture-of-Experts (sliding-window + global attention, FP8 KV cache)
- 提供商
- Poolside
- 许可证
- OpenMDW-1.1
- Practical deployment
- Single consumer GPU
- 规格量化
- Q4_K_M
- 系统内存
- 32 GB
- 记录更新于
- 2026-08-15
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。
| 基准 | 分数 | 来源 |
|---|---|---|
| SWE-Bench Verified | 70.9 / 100 % | 已发布 · https://poolside.ai/blog/introducing-laguna-xs-2-1 |
| SWE-bench Multilingual | 63.1 / 100 % | 已发布 · https://poolside.ai/blog/introducing-laguna-xs-2-1 |
我的 GPU 能运行 Laguna XS 2.1 33B-A3B 吗?
- Poolside Laguna XS 2.1 在 AMD Radeon RX 7800 XT 上
- Poolside Laguna XS 2.1 在 AMD Radeon RX 7900 XT 上
- Poolside Laguna XS 2.1 在 AMD Radeon RX 7900 XTX 上
- Poolside Laguna XS 2.1 在 AMD Radeon RX 9060 XT 16GB 上
- Poolside Laguna XS 2.1 在 AMD Radeon RX 9060 XT 8GB 上
- Poolside Laguna XS 2.1 在 AMD Radeon RX 9070 上
- Poolside Laguna XS 2.1 在 AMD Radeon RX 9070 XT 上
- Poolside Laguna XS 2.1 在 Apple M1 上
- Poolside Laguna XS 2.1 在 Apple M1 Pro 上
- Poolside Laguna XS 2.1 在 Apple M2 上
- Poolside Laguna XS 2.1 在 Apple M2 Pro 上
- Poolside Laguna XS 2.1 在 Apple M3 上
- Poolside Laguna XS 2.1 在 Apple M3 Pro 上
- Poolside Laguna XS 2.1 在 Apple M4 上
- Poolside Laguna XS 2.1 在 Apple M4 Pro 上
- Poolside Laguna XS 2.1 在 Apple M5 上
- Poolside Laguna XS 2.1 在 Intel Arc B570 上
- Poolside Laguna XS 2.1 在 Intel Arc B580 上
- Poolside Laguna XS 2.1 在 NVIDIA GeForce RTX 3060 (12GB) 上
- Poolside Laguna XS 2.1 在 NVIDIA GeForce RTX 3070 上
- Poolside Laguna XS 2.1 在 NVIDIA GeForce RTX 3070 Ti 上
- Poolside Laguna XS 2.1 在 NVIDIA GeForce RTX 3080 (10GB) 上
- Poolside Laguna XS 2.1 在 NVIDIA GeForce RTX 3090 上
- Poolside Laguna XS 2.1 在 NVIDIA GeForce RTX 4060 上
- Poolside Laguna XS 2.1 在 NVIDIA GeForce RTX 4060 Ti 16GB 上
- Poolside Laguna XS 2.1 在 NVIDIA GeForce RTX 4070 上
- Poolside Laguna XS 2.1 在 NVIDIA GeForce RTX 4070 Super 上
- Poolside Laguna XS 2.1 在 NVIDIA GeForce RTX 4070 Ti 上
- Poolside Laguna XS 2.1 在 NVIDIA GeForce RTX 4070 Ti Super 上
- Poolside Laguna XS 2.1 在 NVIDIA GeForce RTX 4080 上
- Poolside Laguna XS 2.1 在 NVIDIA GeForce RTX 4080 Super 上
- Poolside Laguna XS 2.1 在 NVIDIA GeForce RTX 4090 上
- Poolside Laguna XS 2.1 在 NVIDIA GeForce RTX 5060 上
- Poolside Laguna XS 2.1 在 NVIDIA GeForce RTX 5060 Ti 16GB 上
- Poolside Laguna XS 2.1 在 NVIDIA GeForce RTX 5060 Ti 8GB 上
- Poolside Laguna XS 2.1 在 NVIDIA GeForce RTX 5070 上
- Poolside Laguna XS 2.1 在 NVIDIA GeForce RTX 5070 Ti 上
- Poolside Laguna XS 2.1 在 NVIDIA GeForce RTX 5080 上
- Poolside Laguna XS 2.1 在 NVIDIA GeForce RTX 5090 上
- Poolside Laguna XS 2.1 在 NVIDIA L40S 上
- Poolside Laguna XS 2.1 在 NVIDIA RTX 6000 Ada Generation 上
Laguna XS 2.1 33B-A3B — 常见问题
How much VRAM does Laguna XS 2.1 33B-A3B need?
About 21 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 Laguna XS 2.1 33B-A3B run on an RTX 4090 (24 GB)?
Yes. Laguna XS 2.1 33B-A3B needs about 21 GB at Q4_K_M, inside a 24 GB card, at an estimated 171 tokens/sec.
How do I run Laguna XS 2.1 33B-A3B locally?
Install Ollama and run `ollama run poolside-laguna`. That pulls the weights and starts a local OpenAI-compatible endpoint; after the download nothing leaves the machine.