Qwen 3.8 27B available on Cerebras at 1500 tokens/s

	Qwen 3.8 27B available on Cerebras at 1500 tokens/s(inference-docs.cerebras.ai)
	684 points by altertable 6 days ago \| 225 comments
	tl;dr: Cerebras is now serving Qwen 3.8 27B on its public endpoints at ~1500 tokens/s, alongside GPT-OSS 120B at ~3000 tokens/s, with context windows up to 128k/131k on paid tiers. Cerebras notes all public models are unpruned originals, using only selective weight-only quantization for storage while keeping activations, attention, and KV cache at full precision; pruned REAP variants are research-only on Hugging Face.
	HN Discussion: ~Rate limits make the service impractical for real coding tasks despite speed ~High speed comes at significantly higher cost compared to alternatives ↓Local inference alternatives can achieve reasonable speeds without these limitations •Wish Cerebras would expose this model via OpenRouter for broader access ~Output speed is impressive but tool calling and context limits hamper usefulness