All AI models · LLM benchmarks · Methodology

Alibaba

Qwen3.8 27B

Dense 27B vision-language model for coding, agent tasks, and image and video understanding

Model facts

Context window
262K tokens
Maximum output
33K tokens
Cheapest paid input
$0.1 per 1M tokens
Cheapest paid output
$0.4 per 1M tokens
Open weights
yes
Providers
34

Capabilities and modalities

reasoning, tool calling, structured output, attachments, temperature control, open weights, text, image, video, pdf.

Local hardware estimate

At 4K context: Q4_K_M 18.0 GiB, Q8_0 30.9 GiB, F16 55.2 GiB working memory.

Parameters
27.8 billion
Architecture
qwen3_5
Model license
apache-2.0

Planning estimate adapted from the MIT-licensed whichllm estimator using metadata from Hugging Face; not a vendor minimum requirement.

Artificial Analysis benchmarks

Compare this model on the full LLM benchmark leaderboard.

ConfigurationAAICodingMathOutput tokens/s
xhigh52.068.1not measured42.9
medium44.556.1not measured44.8
low42.958.2not measured44.2
Non-reasoning34.744.6not measured47.6

Observed price history

  • 2026-08-24: $0.1 input / $0.4 output per 1M tokens

API providers