Qwen3.5 397B-A17B vs Qwen3.8 27B vs GLM-5

It is close. Our weighted score puts them within 2 points (Qwen3.8 27B 67/100, Qwen3.5 397B-A17B 65/100, GLM-5 55/100), so choose by what matters most for your work: Qwen3.8 27B for raw capability. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.

Qwen3.8 27B is cheaper at $0.40 input / $2.50 output per million tokens (median across 39 API providers). Qwen3.5 397B-A17B costs $0.60 input / $3.60 output per million tokens (official Alibaba API price); GLM-5 costs $1.00 input / $3.20 output per million tokens (official Z.AI API price). At a typical mix of three input tokens to one output token, that is $0.925 per million tokens for Qwen3.8 27B versus $1.35 for Qwen3.5 397B-A17B (1.5× as much) and $1.55 for GLM-5 (1.7× as much).

Qwen3.8 27B scores higher on the Capabilities Index (ECI): Qwen3.8 27B 149.4 (#53 of 148), Qwen3.5 397B-A17B 146.7 (#67 of 148) and GLM-5 145.8 (#74 of 148). The confidence ranges of the top two overlap (147.5–151.6 vs 144.8–148.2), so treat the gap as small.

There are no published SWE-bench Verified results for Qwen3.5 397B-A17B and Qwen3.8 27B yet, so there is no like-for-like coding score. On overall capability, Qwen3.8 27B leads, which tends to carry over to coding, but test on your own codebase. All three support tool calling for agent workflows.

Qwen3.5 397B-A17B and Qwen3.8 27B have the largest context windows (262,144 and 262,144 tokens), against 204,800 for GLM-5. Maximum output per response: Qwen3.5 397B-A17B up to 65,536, Qwen3.8 27B up to 32,768, GLM-5 up to 131,072 tokens.

Qwen3.5 397B-A17B accepts text, images, audio and video; Qwen3.8 27B accepts text, images and video; GLM-5 accepts text. Qwen3.5 397B-A17B handles the widest range of inputs.

Yes, all three publish their weights, so you can self-host them.

Qwen3.8 27B is the newest, released Aug 14, 2026. Qwen3.5 397B-A17B came out Feb 15, 2026; GLM-5 came out Feb 12, 2026.

Each model gets a 0–100 score on capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Oct 4, 2026.