B

Mausbiber's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

New activity in cyankiwi/Ministral-3-14B-Reasoning-2512-AWQ-4bit 29 minutes ago

Benchmarked vs AWQ 8bit on 2x RTX 5060 Ti (32GB, vLLM 0.29): 97.7% of 8bit score with 5.4 GiB less VRAM

#1 opened 29 minutes ago by

Mausbiber

New activity in cyankiwi/Ministral-3-14B-Reasoning-2512-AWQ-8bit about 3 hours ago

Benchmarked vs AWQ 4bit on 2x RTX 5060 Ti (32GB, vLLM 0.29): 34.6% overall, +1.5% on SimpleQA Verified

#1 opened about 3 hours ago by

Mausbiber

New activity in mistralai/Ministral-3-14B-Reasoning-2512 about 5 hours ago

AWQ 8bit vs AWQ 4bit local GPU benchmark on 2x RTX 5060 Ti (32GB, vLLM 0.29)

#10 opened about 5 hours ago by

Mausbiber

New activity in cosmicproc/Qwen3.5-4B-NVFP4 about 6 hours ago

Benchmarked vs BF16 on RTX 5060 Ti (vLLM 0.29): 263.2 tok/s throughput and full accuracy comparison

❤️ 1

#1 opened about 6 hours ago by

Mausbiber

New activity in xbill9/gemma-4-E4B-it-qat-q4_0-w4a16-ct-text-ple4 1 day ago

Benchmarked vs BF16 on 2x RTX 5060 Ti (vLLM 0.29): 219.2 tok/s, 21m runtime, matches BF16 on IFBench & SciCode

#1 opened 1 day ago by

Mausbiber

New activity in zai-org/GLM-4.7-Flash 1 day ago

Local GPU benchmark results on 2x RTX 5060 Ti (32GB, vLLM 0.29): 139.0 tok/s and 49.1% Thinking score

#76 opened 1 day ago by

Mausbiber

New activity in Qwen/Qwen3.6-35B-A3B 1 day ago

Local GPU benchmark on 2x RTX 5060 Ti (32GB): 210.2 tok/s and #1 on SimpleQA Verified (19.4%)

#80 opened 1 day ago by

Mausbiber

New activity in Qwen/Qwen3.6-27B 1 day ago

Local GPU benchmark on 2x RTX 5060 Ti (32GB): 36.9% SciCode and comparison vs Qwen3.8 27B & Qwen3.5 27B

#55 opened 1 day ago by

Mausbiber

New activity in Qwen/Qwen3.8-27B 1 day ago

Local GPU benchmark results: Qwen3.8 27B vs Qwen3.6 27B vs Qwen3.5 27B on 2x RTX 5060 Ti (32GB)

#225 opened 1 day ago by

Mausbiber

New activity in cyankiwi/Qwen3.5-27B-AWQ-4bit 1 day ago

Full benchmark results on 2x RTX 5060 Ti (32GB, vLLM 0.29): #1 in Non Thinking and #2 in Thinking mode

#11 opened 1 day ago by

Mausbiber

New activity in google/gemma-4-E4B-it 1 day ago

BF16 vs 4bit QAT benchmark on RTX 5060 Ti (vLLM 0.29): 32 min runtime and almost zero quality loss in 4bit

#55 opened 1 day ago by

Mausbiber

New activity in Qwen/Qwen3.5-4B 1 day ago

BF16 vs NVFP4 quantization benchmark on RTX 5060 Ti (vLLM 0.29): How much quality does 4bit cost on a 4B model?

#33 opened 1 day ago by

Mausbiber

New activity in QuantTrio/GLM-4.7-Flash-AWQ 1 day ago

Benchmarked on 2x RTX 5060 Ti (32GB, vLLM 0.29 awq_marlin): 139.0 tok/s, 85.6% IFEval

#3 opened 1 day ago by

Mausbiber

New activity in unsloth/Qwen3.6-27B-NVFP4 1 day ago

Benchmarked on 2x RTX 5060 Ti (32GB, vLLM 0.29): #1 on ARC AGI 1 (12.9%) and 36.9% on SciCode

#17 opened 1 day ago by

Mausbiber

New activity in nvidia/Qwen3.6-35B-A3B-NVFP4 1 day ago

Benchmarked on 2x RTX 5060 Ti (32GB, vLLM 0.29): 210.2 tok/s, 54.1% overall in just 3.1 hours

#19 opened 1 day ago by

Mausbiber

New activity in unsloth/Qwen3.8-27B-NVFP4 1 day ago

Benchmarked on 2x RTX 5060 Ti (32GB, vLLM 0.29): 69.8% IFBench, 92.8% IFEval at 60.7 tok/s

#27 opened 1 day ago by

Mausbiber

New activity in nvidia/Gemma-4-26B-A4B-NVFP4 1 day ago

Benchmarked on 2x RTX 5060 Ti (32GB) with vLLM 0.29: #1 overall score at 151.1 tok/s

#8 opened 1 day ago by

Mausbiber

New activity in google/gemma-4-26B-A4B-it 1 day ago

Local GPU benchmark on 2x RTX 5060 Ti (32GB, vLLM 0.29): #1 overall score (57.9%) at 151.1 tok/s

#69 opened 3 days ago by

Mausbiber

New activity in Qwen/Qwen3.5-27B 1 day ago

Local GPU benchmark on 2x RTX 5060 Ti (32GB, vLLM 0.29): #1 in Thinking Off (31.8%) and #2 in Thinking On (57.1%)

#53 opened 3 days ago by

Mausbiber

New activity in nvidia/Qwen3.6-35B-A3B-NVFP4 1 day ago

Benchmarked on 2x RTX 5060 Ti (32GB, vLLM 0.29): 210.2 tok/s, 54.1% overall in just 3.1 hours

#19 opened 1 day ago by

Mausbiber