← LLM

Cerebras

Qwen3.8-27B

gateway · reasoning low (gateway default)

Measured 2026-08-25

* Conditional

Same weights as the Alibaba Qwen3.8-27B row on a different host and at a different reasoning setting, so read the pair as host+setting, not host alone.

Score

No real-time first-token measurement, so a speed-weighted score is not computed — unranked rather than judged on partial data.

Task done

87.72%

76–96% 95% CI bootstrap over items · n=95 · measured 2026-09-06

Refusal quality

80.0%

names the gap 70% · offers a route 83% · 10 probes x 3 iterations · 2 silent turns scored zero

Fabrication

0%

0 of 10 probes (30 probe runs) · ≤31%

Dead-air

33.2%

98 of 295 turns

Tool silence

48.7%

97 of 199 tool-call turns

Stalled

14.4%

13 of 90 runs

TTFT p50

Not measured through the gateway — this column times the Speko hop and this row was measured vendor-direct, so a number from it would not be comparable to the other rows. The vendor-direct regions under Region carry this row’s first-token number: 103ms p50 in US East and 329ms in Singapore, n=80 each, at the gateway’s own reasoning-low clamp.

Cost / 1M tok

$1.49

Cerebras list pricing for this model, $0.99/1M input and $1.49/1M output, from the vendor’s own model page (inference-docs.cerebras.ai/models/qwen-3.8-27b, read 2026-09-06). The same weights cost $2.55/1M output on the Alibaba row.

Share this result

Embed the live badge

Speko llm rank
Markdown
[![Speko llm rank](https://benchmarks.speko.ai/badge/llm/cerebras-qwen3-8-27b.svg)](https://benchmarks.speko.ai/llm/cerebras-qwen3-8-27b)
HTML
<a href="https://benchmarks.speko.ai/llm/cerebras-qwen3-8-27b"><img src="https://benchmarks.speko.ai/badge/llm/cerebras-qwen3-8-27b.svg" alt="Speko llm rank"></a>
URL
https://benchmarks.speko.ai/llm/cerebras-qwen3-8-27b