Is DeepSeek-V3.2-Exp out yet?
Yes.
About DeepSeek-V3.2-Exp
Experimental sparse-attention release that halved API prices.
How we confirmed it
Released via official access
September 29, 2025 at 12:00 AM UTC
Release timeline
Waiting for DeepSeek's next model?
One click creates your free account (or signs you into an existing one) and turns on alerts for DeepSeek releases. There's no password to set, and you can unsubscribe any time.
Benchmarks
Agentic#35
DeepSeek V3.2 · via BenchLM.ai · captured Jul 14, 2026 · source
55.8
Knowledge#40
DeepSeek V3.2 (Thinking) · via BenchLM.ai · captured Jul 17, 2026 · source
64.1
Math#49
DeepSeek V3.2 (Thinking) · via BenchLM.ai · captured Jul 3, 2026 · source
51.6
Multilingual#46
DeepSeek V3.2 (Thinking) · via BenchLM.ai · captured Jul 17, 2026 · source
61.6
Reasoning#45
DeepSeek V3.2 (Thinking) · via BenchLM.ai · captured Jul 17, 2026 · source
53.4
LM Arena#108
deepseek-v3.2-exp · via LM Arena · 11,987 votes · captured Aug 16, 2026 · source
1422.5666789881102
Coding#50
DeepSeek V3.2 (Thinking) · via BenchLM.ai · captured Jul 17, 2026 · source
52.32
MultimodalGrounded#43
DeepSeek V3.2 (Thinking) · via BenchLM.ai · captured Jul 14, 2026 · source
59.2
Sourced metrics mirrored with attribution - not community sentiment.
Data from BenchLM.ai.
Vibe rating
How this model actually feels to use, rated by the people here.
No vibe ratings yet. Be the first.
Sign in to rate this model.
Model Pulse
How is this model behaving today? One check-in per person per day: the running record of “did they nerf it”. Resets at midnight UTC.
No check-ins yet today. Be the first to log how this model is behaving.
Community consensus
An AI summary of the public comments here. Not a sourced fact.
Not enough public feedback yet to summarize. Add a comment to help build the picture.
Compare DeepSeek-V3.2-Exp with another model →
Also waiting on: Gemini 3.5 Pro, Claude Fable 5.1, Claude Mythos 5.1, GLM-6
Comments
Missing a model, or wishing OutYet did something it doesn't? Tell us what to build.