Qwen3.5 0.8B

Qwen·Pending·Released 2026-03-02textvisionvideo
Context262.144k
Max output-
Input / 1M-
Output / 1M-
Cached in-
Benchmarks14

Benchmarks

Cited public sources · rank is exact-benchmark, apples-to-apples

BFCL-V4 (thinking)#2/225.3% ↗
IFBench (thinking)#2/221% ↗
IFEval (non-thinking)#2/252.1% ↗
IFEval (thinking)#2/244% ↗
LongBench v2 (thinking)#2/226.1% ↗
MMMLU (thinking)#2/244.3% ↗
MMMU (thinking)#2/249% ↗
MMMU-Pro (thinking)#2/231.2% ↗
TAU2-Bench (thinking)#2/211.6% ↗

Qwen3.5 0.8B FAQ

What is Qwen3.5 0.8B?
Qwen3.5 0.8B is a large language model from Qwen, released on March 2, 2026. It accepts text, image and video input.
When was Qwen3.5 0.8B released?
Qwen released Qwen3.5 0.8B on March 2, 2026.
What is Qwen3.5 0.8B's context window?
Qwen3.5 0.8B has a 262K-token context window.
How does Qwen3.5 0.8B score on benchmarks?
Qwen3.5 0.8B has 14 cited benchmark scores, including BFCL-V4 (thinking) 25.3%, GPQA (thinking) 11.9% and IFBench (thinking) 21%. Each score links to its source.

Access Qwen3.5 0.8B and every other model through one endpoint with automatic failover: Respan gateway.