Qwen 3.8 27B 现已在 Cerebras 上线，速度为每秒 1500 token。

Qwen 3.8 27B 现已在 Cerebras 上线，速度为每秒 1500 token。
Qwen 3.8 27B available on Cerebras at 1500 tokens/s

原始链接: https://inference-docs.cerebras.ai/models/overview

Cerebras 通过其公共 API 提供一系列开源模型，用户可通过免费试用或按需付费层级进行访问。对于高吞吐量、生产级别的需求，用户可选择专用端点（Dedicated Endpoints）。关于模型完整性与压缩，Cerebras 始终保持完全透明： * **无剪枝（No Pruning）：** 所有公共 API 模型均为未经剪枝的原始版本。任何未来对剪枝技术的探索都将托管在独立且标注清晰的端点上，以确保用户知情。 * **存储优化：** 虽然权重通过选择性量化（例如 4 位到 16 位）来节省空间，但公司采用实时反量化技术，确保运算以高精度执行。激活值、注意力机制及 KV 缓存保持完全的非量化精度，以维持最高质量。 * **研究访问：** 内部研究（如 REAP）产生的剪枝模型仅发布在 Hugging Face 上供实验使用，不通过生产 API 提供。用户可查阅模型选择指南，为特定使用场景寻找合适方案，或按照快速入门指南开始进行 API 调用。

Cerebras 近期上线了 Qwen 3.8 27B 模型，并实现了每秒 1,500 token 的推理速度，这引发了关于该平台在实际编程和智能体工作负载中实用性的讨论。尽管其原始速度前所未有，但用户反映存在显著的实际阻碍。主要的痛点在于 Cerebras 设定的每分钟 150,000 至 450,000 token (TPM) 的限制，且该限额包含了输入和缓存的 token。由于智能体任务频繁需要大上下文窗口和反复的工具调用，用户往往在几秒钟内就会触及上限，导致立即被限流。此外，由于该服务对缓存的 token 缺乏有效的价格折扣，使得其在长时间编程任务中的成本远高于标准云服务提供商。用户还指出，Cerebras 在支持和计费流程上存在明显摩擦，尤其是对于非企业级客户而言。许多开发者认为，虽然该技术作为“速度演示”令人印象深刻，但严格的速率限制、高昂的成本以及 128k 上下文窗口的约束，使其难以满足生产级编程的需求。因此，许多开发者正转向本地推理、优化硬件或其他能为复杂、持续任务提供更具可持续性性能的平台。

Models on Cerebras public endpoints are available on the free trial and pay-as-you-go tiers, subject to rate limits and pricing. For additional model families, reserved capacity, higher throughput, and production SLAs, see Dedicated Endpoints.

New here? Follow the Quickstart to make your first API call. To pick a model by use case, see the model selection guide. Select any model name below for full specs, capabilities, and per-tier limits.

Available Models

Looking for more models? Many additional model families are available through Dedicated Endpoints.

Model Compression

This section provides transparency about the compression state of each model available on our platform. We host a variety of open-source models from the community. We do not currently host pruned models on our public endpoints. All models served through our public endpoints are the original, unpruned versions. While we conduct research on pruning techniques like REAP (Router-weighted Expert Activation Pruning), these pruned models are shared with the research community on Hugging Face but are not available through our shared API. You can read more about REAP in our research blog. All of our public models are unpruned. Cerebras uses selective weight-only quantization only during storage to preserve maximal quality. This means that the weights are stored in partial 16-bit / 8-bit / 4-bit, in-line with industry standards. For quality, sensitive layers are stored at full precision with dequantization on the fly, so operations are done in high precision. The activations, attention, and kv cache remain in full precision and unquantized.

Frequently Asked Questions

Will you change a model's architecture without notice?

No. We are committed to serving the original models for all existing endpoints, without modification. We do not alter model architectures via pruning on our hosted portfolio. If we explore additional compression techniques (like pruning) in the future, these would be offered as separate endpoints with pruning-specific names, ensuring complete transparency and allowing you to choose which version best fits your needs.

Where can I find your REAP pruned models?

Our REAP pruned models are available on Hugging Face for research and experimentation purposes: Cerebras REAP Collection. These models demonstrate our pruning research but are not served through our production API.

What are compression, quantization, and pruning?

Compression is an umbrella term for techniques that reduce model size or computational requirements. Common compression techniques include:

Quantization: Reducing the precision of numbers used to represent model weights (e.g., converting from FP16 to FP8). This reduces memory usage without changing the model’s architecture.
Pruning: Permanently removing parts of a model, like layers or experts, to reduce model size. This changes the model’s architecture and creates a different model.