ghostplant
ghostplant
AI & ML interests
None yet
Recent Activity
new activity about 9 hours ago
lukealonso/GLM-5.2-NVFP4:launch failed with EAGLE new activity about 9 hours ago
lukealonso/GLM-5.1-NVFP4:Tested Docker working for Amphere/Hopper/Blackwell/MI300 new activity 4 days ago
lukealonso/GLM-5.2-NVFP4:Have you tested the AIME-2026 scores for this version?Organizations
None yet
launch failed with EAGLE
➕ 1
2
#3 opened 2 days ago
by
luchangli03
Tested Docker working for Amphere/Hopper/Blackwell/MI300
#7 opened 2 months ago
by
ghostplant
Have you tested the AIME-2026 scores for this version?
👍 1
#2 opened 4 days ago
by
ghostplant
Expecting Benchmark-validated GLM-5.2-NVFP4
🚀 4
#4 opened 4 days ago
by
ghostplant
Is AIME 2026 Benchmark evaluated with Tool Call enabled?
👍 1
1
#14 opened 5 days ago
by
ghostplant
Model doesn't call tools
1
#3 opened 15 days ago
by
infinitebibliography
Guide to run Kimi K2.6 on CPU, GPU and SSD setups! 🔥
👍 1
2
#17 opened 2 months ago
by
danielhanchen
Cannot reproduce claimed 96.3 AIME 2025 score
1
#7 opened 3 months ago
by
NonoRiri-7
Kimi-K2.6 support
2
#8 opened 2 months ago
by
g-a-b-y
Expecting New NVFP4 for GLM-5.1 ASAP, Thanks!
👀 1
11
#6 opened 3 months ago
by
ghostplant
Greater than Opus, GLM-5.x is amazing! We start to help this model on A100/MI300.
1
#74 opened 2 months ago
by
ghostplant
Please support nvidia/GLM-5-NVFP4
#2 opened 3 months ago
by
ghostplant
For Pre-blackwell Architecture (A100/H100)
#4 opened 4 months ago
by
ghostplant
DSA Question
1
#33 opened 6 months ago
by
ghostplant
For those who need a simplified execution on NVIDIA GPU
🔥 1
#21 opened 6 months ago
by
ghostplant
Question about long-context evaluation in DeepSeek-V3.2-Exp
1
#15 opened 9 months ago
by
fcMpKYz6Avp5QK
Can gpt-oss support local vllm deployment on a100 GPU?
10
#73 opened 11 months ago
by
Cola-any
Running gpt-oss Without FlashAttention 3 – Any Alternatives to Ollama?
3
#72 opened 11 months ago
by
shinho0902
Run GPT-OSS-120B with just Single A100 (80GB)
2
#80 opened 11 months ago
by
ghostplant
How is Qwen3's inv_freq computed from scratch?
#13 opened 11 months ago
by
ghostplant