Search results

GLM-4.7 Browsecomp/HLE Reproducibility | 结果复现

Status: Open.

#120 In zai-org/GLM-4.5;

· panademo opened on Dec 23, 2025

The scores for K2-thinking and MiMo-V2-Flash were mixed up

Status: Open.

#118 In zai-org/GLM-4.5;

· yetlinghao opened on Dec 23, 2025

GLM4.7 在openhand 表现不佳

Status: Open.

#117 In zai-org/GLM-4.5;

· NebulaMao opened on Dec 23, 2025

Always reporting "The following fields were present in the request but ignored: {'reasoning'}" when using claude code with claude-code-router

Status: Open.

#112 In zai-org/GLM-4.5;

· ThomasLWang opened on Dec 7, 2025

sglang部署后，请求4.6无法禁止think

Status: Open.

#110 In zai-org/GLM-4.5;

· hebly723 opened on Nov 25, 2025

GLM 预测第三个token的接受率基本为0的原因

Status: Open.

#109 In zai-org/GLM-4.5;

· echomingzhu opened on Nov 24, 2025

GLM4.6 Context at 104k

duplicateThis issue or pull request already existsThis issue or pull request already exists

Status: Open.

#108 In zai-org/GLM-4.5;

· sant3e opened on Nov 12, 2025

vLLM or SGLang Command Line for Maximum Throughput?

Status: Open.

#107 In zai-org/GLM-4.5;

· seungduk-yanolja opened on Nov 12, 2025

加载模型警告： Some weights of the model checkpoint at /model_path were not used when initializing Glm4MoeForCausalLM

Status: Open.

#106 In zai-org/GLM-4.5;

· Minami-su opened on Nov 10, 2025

CUDA Error Happens when using vLLM in inference: an illegal memory access was encountered.

Status: Open.

#105 In zai-org/GLM-4.5;

· AkaTsukijm opened on Nov 6, 2025

GLM-4.6 Many times Fails to Maintain Conversation Context, Ignores Prior Discussion on Model Calibration and Responds Off-Topic

Status: Open.

#104 In zai-org/GLM-4.5;

· rushabhT3 opened on Nov 5, 2025

Potential security issue: unrestricted access to hidden models and unlimited API use via frontend tokens

Status: Open.

#103 In zai-org/GLM-4.5;

· Akzestia opened on Nov 3, 2025