Artificial Analysis lists Qwen3.5 0.8B
Artificial Analysis has published a benchmark page for Qwen3.5 0.8B, a two-model release that will be compared against other open-weights models of 4B parameters or fewer.
Independent benchmarking firm Artificial Analysis has added a release page for Qwen3.5 0.8B. The page says the release "offers 2 models," each with different intelligence, performance and pricing. The text we have does not say how the two variants differ.
The page sets out how the models will be compared. Open-weights models are grouped by parameter count: Tiny (≤4B), Small (4B–40B), Medium (40B–150B) and Large (>150B). At 0.8B parameters, Qwen3.5 0.8B falls in the Tiny class. Reasoning models are compared against both reasoning and non-reasoning models. Non-reasoning models are compared only with each other.
The page has charts for its Intelligence Index against cost per task, cost per Intelligence Index task, and time per task. It also shows several capability indexes, each built from a set of evaluations. These include AA-Briefcase v1.1, AA-Omniscience v1.1, HLE-Diamond, AutomationBench-AA, GDP.pdf, MLCR-AA, CritPt v2 (Hard Subset), Terminal-Bench-Science 0.1, Terminal-Bench 4.0 and FrontierCode-AA 2.0.
The text we have includes no scores, prices or speeds for either model, so we can't yet say how Qwen3.5 0.8B compares with other tiny open-weights models. It also gives no details on the release itself, such as its date, license or architecture.