readable AI benchmarks

Alibaba

Qwen3.8 27B (Non-reasoning)

non-reasoning · open weights · Aug 14, 2026

Qwen3.8 27B (Non-reasoning) has a Quality Score of 32.4, ranking 88th among 259 scored models. Per 1M tokens, pricing is $0.500 input, $3.00 output, and $0.050 cached input. It accepts text, image, and video input, outputs text, and has a 256k token context window. Its Reliability Score is 46.0, ranking 62nd among 259 scored models, above average (total average: 33.1). Its Value Score is 27.3, around average compared with the total Quality average of 26.6.

Quality Score32.4
Value Score27.3
Reliability46.0
Cache Discount90%

Model specification

Reasoning
non-reasoning
Input modalities
text, image, video
Output modalities
text
Context window
256k tokens
Weights
open weights

Token prices USD per 1M tokens

Input
$0.500
Output
$3.00
Cached Input
$0.050
Cached Output
not available
Cost per task
$0.214

Capability

Intelligence
34.7
Coding
44.6
Agentic
30.4
Omniscience
-8.0
Correct
8.6%
Blended price
$0.435

Answer outcomes

Correct 8.6%Incorrect 16.5%Abstained 74.9%

Artificial Analysis benchmarks

GDPval-AA v2
36%
τ³-Banking
20%
Terminal-Bench v2.1
49%
SciCode
36%
Humanity’s Last Exam
12%
GPQA Diamond
82%
CritPt
0%
AA-Omniscience
46%
AA-LCR
63%

Similar models