Qwen3.8 27B Xhigh vs Beam 1.0: speed and benchmark comparison

If you only looked at the intelligence index, Qwen3.8 27B (Xhigh) and Beam 1.0 would be hard to tell apart. Beam 1.0 streams output at about 344 tokens per second, 9 times the rate of Qwen3.8 27B Xhigh. All numbers are the catalogue's recorded values; nobody ran either model by hand for this page.

Beam 1.0 is the faster writer

Once a response is under way, Qwen3.8 27B Xhigh produces roughly 38 tokens per second against 344 tokens per second for Beam 1.0. On long outputs, that is a real advantage for Beam 1.0. Time to first token is 4.1 seconds for Qwen3.8 27B Xhigh and 0.70 seconds for Beam 1.0. Beam 1.0 gets to its first token 5.8 times as fast.

For a 600-token response, first token plus generation comes to roughly 20 seconds on Qwen3.8 27B Xhigh and 2.4 seconds on Beam 1.0.

Generation gap

Beam 1.0 arrived on October 16, 2026, about 2 months after Qwen3.8 27B Xhigh (August 14, 2026).

Price per token

Per million tokens, Qwen3.8 27B Xhigh charges $0.50 for input and $3.00 for output, while Beam 1.0 charges $0.40 and $1.20. For output tokens, Qwen3.8 27B Xhigh is 2.5 times the price of Beam 1.0.

For a sample workload of 1 million input tokens and 200,000 output tokens, Qwen3.8 27B Xhigh would cost $1.10 and Beam 1.0 $0.64, a difference of $0.46. A bigger example, a month of coding-assistant traffic at 30 million input and 3 million output tokens, costs $24.00 on Qwen3.8 27B Xhigh and $15.60 on Beam 1.0: a gap of $8.40. Blended at three input tokens per output token, the effective rates are $1.125 for Qwen3.8 27B Xhigh and $0.60 for Beam 1.0.

Qwen3.8 27B Xhigh leads on the intelligence index

Qwen3.8 27B Xhigh records 20.1 on the intelligence index and Beam 1.0 records 17.7. The lead is visible but narrow, so it should not decide the choice on its own. Put another way, Beam 1.0 would need 14% more on this index to draw level.

For context, that places Qwen3.8 27B Xhigh 78th and Beam 1.0 84th out of 109 scored models still on offer.

For coding or maths work specifically, the catalogue has no dedicated index on either side yet.

Cost per point of intelligence

A blunt value measure, blended price over intelligence score, puts Qwen3.8 27B Xhigh at $0.056 and Beam 1.0 at $0.034 per point. Beam 1.0 delivers each index point for 39% less; whether Qwen3.8 27B Xhigh's extra 2.4 points are worth the premium depends on how often your tasks need them.