GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
salamanderDixon
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
salamanderDixon
Model Tree
Qwen/Qwen3.6-27B
Input Modalities
Output Modalities
Supported Functionality
| Parameter | Value |
|---|---|
| direction_index | 41.77 |
| attn.o_proj.max_weight | 1.22 |
| attn.o_proj.max_weight_position | 51.32 |
| attn.o_proj.min_weight | 1.21 |
| attn.o_proj.min_weight_distance | 32.79 |
| mlp.down_proj.max_weight | 1.44 |
| mlp.down_proj.max_weight_position | 42.02 |
| mlp.down_proj.min_weight | 0.13 |
| mlp.down_proj.min_weight_distance | 37.15 |
| Metric | This model | Original model (TeichAI/Qwen3.6-27B-Claude-Opus-Reasoning-Distill-v2) |
|---|---|---|
| KL divergence | 0.0774 | 0 (by definition) |
| Refusals | 8/100 | 98/100 |
text
Qwen3.6-27B-Claude-Opus-Reasoning-Distill-v2 arc arc/e boolq hswag obkqa piqa winomxfp8 0.665,0.831,0.910,0.790,0.456,0.813,0.772 Qwen3.6-27B arc arc/e boolq hswag obkqa piqa winomxfp8 0.647,0.803,0.910,0.773,0.450,0.806,0.742Provided by @nightmedia. All benchmarks were done in mxfp8 precision
If you need help setting up and configuring this model please follow the Qwen team's instructions in the original model's README
Qwen3.6-27B-Claude-Opus-Reasoning-Distill-v2 arc arc/e boolq hswag obkqa piqa winomxfp8 0.665,0.831,0.910,0.790,0.456,0.813,0.772Qwen3.6-27B arc arc/e boolq hswag obkqa piqa winomxfp8 0.647,0.803,0.910,0.773,0.450,0.806,0.742