GLM vs Qwen3 for local use: which top open model to actually run
Qwen3-30B-A3B fits a 24GB card at ~17.5GB; GLM-4.5-Air needs ~60GB. GLM ranks a touch higher; Qwen3 runs on far less. How to choose, with sourced…
Aliteq
Lena Fischer · AI & Local Compute Editor
The short answer
For running locally, Qwen3-30B-A3B (30B total / 3B active) fits in about 17.5GB — it runs on a single 24GB consumer card. GLM-4.5-Air (106B / 12B active) needs about 60GB at 4-bit, so it wants an…
Qwen3-30B-A3B: 30B / 3B active, ~17.5GB — fits a single 24GB card.
GLM-4.5-Air: 106B / 12B active, ~60GB at 4-bit — wants 80GB or 128GB unified.
Quality: GLM-5.3 ties Kimi K3 at the top of open-weight rankings; Qwen3 is close and lighter.
Both are MoE with small active-parameter counts, so both are fast for their size.
Rule of thumb: modest hardware → Qwen3; more memory + top quality → GLM.
Aliteq
Read the full story
GLM vs Qwen3 for local use: which top open model to actually run