Unverified50% confidenceBenchmarkExact time

Qwen3 27B's performance on coding and agent benchmarks is: SWE-bench Pro 61.7, Terminal-Bench 2.1 73.0, and LiveCodeBench V6 90.3.

1

Sources

50%

Confidence

Medium-term (~90 days)

Relevance

9/6/2026

First Seen

Valid until: 12/5/2026

Sources

Qwen3 27B Open-Sourced: A Multimodal Agent Model That Runs on a Single GPU

Related Entities

Qwen3 27Bmodel SWE-bench Proconcept Terminal-Benchconcept LiveCodeBenchconcept

Related Claims

UnverifiedThe Qwen 3.5-32B model is available on the Bailian platform.67% similar UnverifiedQwen3.6-27B uses a Dense architecture rather than a MoE (Mixture of Experts) architecture.67% similar UnverifiedSeries such as Qwen2.5 and Gemma3 have already released 1B–3B quantized versions optimized specifically for edge scenarios.66% similar UnverifiedThe Qwen3-32B model contains 64 Transformer decoder layers.66% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/863356

API

curl https://kongchang.com/api/v1/knowledge/claims/863356

MCP

get_claim(id=863356)