Qwen3.8-Whittle-MoE-27B-MLX-4bit

barozp/Qwen3.8-Whittle-MoE-27B-MLX-4bit

Text Generation26.9B parametersmlx-lmLicense: apache-2.0
Tracked by OpenModelStats since August 27, 2026View on Hugging Face ↗
Downloads 30D
2,540
Total downloads
3,377
Gained 7D
+508(increase)
20.1%(up) vs baseline
Gained 30D
—
Likes
5
+0 in 7D
Rank · tracked models
#17,217
990(down 990 places)this week
Parameters
26.9B
Last source update
Aug 24, 2026
Momentum
58.6
  • Qwen3.8-Whittle-MoE-27B-MLX-4bit gained 508 downloads during the last seven tracked days.
  • It moved from #4180 to #4455 among tracked Text Generation models this week.

Downloads over time

Daily gain between consecutive observations

Likes over time

Rank among tracked models

Lower is better

Overview

Qwen3.8-Whittle-MoE-27B-MLX-4bit is a text generation model published by barozp. OpenModelStats has tracked the model since Aug 27, 2026, most recently observing it Sep 26, 2026.

Published
Aug 24, 2026
Last updated
Aug 24, 2026
Library
mlx-lm
License
apache-2.0
Task
Text Generation
Parameters
26,917,293,056
Spaces
—
Derivative models
—
Velocity 7D/day
73
mlx-lmsafetensorsqwen3_5_moemlxqwenqwen3reasoningopus-distillwhittleexpert-pruningmoerouter-healing

Publisher

barozpView publisher statistics →

Derived from logic65/Qwen3.8-Whittle-MoE-27B-A17.8B (reported by the source; parent not yet tracked).

Similar models

Models similar to Qwen3.8-Whittle-MoE-27B-MLX-4bit
ModelDL 30D
Qwen3.8-27B-Ternary-Bonsai-2-DFlash2-MLXnathansutton5,644
mnma_qwen3.8_27b_nvfp4minima-ai933
Hemmingway-1Altworld5,590
Qwen3.6-27B-OBLITERATEDOBLITERATUS5,240
Qwen3.8-27B-abliterated-AWQ-INT4hotdogs1,792
Qwen3.8-27B-Obliterated-E03manitcor1,565
barozp/Qwen3.8-Whittle-MoE-27B-MLX-4bit — Downloads, Growth & Stats | OpenModelStats