jashepp/

Equinox-31B-i1-MXFP4-GGUF

💎 Equinox-31B - Custom Mixed Precision GGUFs with Imatrix

Equinox draws its name from the balance between extremes. Trained on a balanced blend of Wayfarer 2's unforgiving dark adventures and Hearthfire's quiet slice-of-life storytelling, Equinox is equally at home in perilous dungeons and candlelit conversations.
Equinox 31B was trained with two epochs of SFT (Supervised Fine-Tuning) on top of Gemma 4 31B Instruct.

📄 GGUF Files

• Used larger datasets for imatrix
• Added Equinox-31B-MXFP4-Only-Imatrix.gguf

• Used a new combined/blended imatrix
• MXFP4_Q8_0 & MXFP4_Q6_K
  • Renamed to use model-quants-imatrix style naming
  • Corrected --tensor-type regex

📊 Importance Matrix (Imatrix)

🔍 Precision Matrix & Flavor Variations

1. The Tri-Quant Hybrid Flavor (MXFP4 + Q8_0 + F16)

• Tier 1 (Core - F16 Precision): token_embd.weight - Protects the critical input/output vocabulary mappings. Dramatically prevents text degradation.
• Tier 2 (Backbone - Q8_0 Precision): attn_q, attn_k, attn_v, attn_output, attn_qkv - All core attention layers. Keeps the model's primary attention mechanism high quality.
• Tier 3 (Dense FFN - MXFP4 Precision): ffn_down, ffn_gate, ffn_up - Downsamples the massive, heavy feed-forward network layers to hardware-optimized 4-bit microscaling blocks.

2. The Dual-Quant Hybrid Flavor (MXFP4 + Q8_0)

• Tier 1 (Backbone - Q8_0 Precision): Vocabulary embeddings (token_embd), final logit projections (output), layer normalizations, and structural layers fall back to the heavy Q8_0 format.
• Tier 2 (Compute Layers - MXFP4 Precision): attn_k, attn_q, attn_output, ffn_down, ffn_gate, ffn_up - The core attention projections and dense feed-forward network blocks are target-quantized directly to MXFP4 to leverage highly optimized hardware kernels.

3. The Tri-Quant Hybrid Flavor (MXFP4 + Q6_K + Q8_0)

• Tier 1 (Core - Q8_0 Precision): token_embd.weight - Keeps the critical input/output vocabulary mappings. Lessens the text degradation.
• Tier 2 (Backbone - Q6_K Precision): attn_q, attn_k, attn_v, attn_output, attn_qkv - All core attention layers. Keeps the model's primary attention mechanism at decent quality.
• Tier 3 (Dense FFN - MXFP4 Precision): ffn_down, ffn_gate, ffn_up - Downsamples the massive, heavy feed-forward network layers to hardware-optimized 4-bit microscaling blocks.

4. The Dual-Quant Hybrid Flavor (MXFP4 + Q6_K)

• Tier 1 (Backbone - Q6_K Precision): Vocabulary embeddings (token_embd), final logit projections (output), layer normalizations, and structural layers fall back to the heavy Q6_K format.
• Tier 2 (Compute Layers - MXFP4 Precision): attn_k, attn_q, attn_output, ffn_down, ffn_gate, ffn_up - The core attention projections and dense feed-forward network blocks are target-quantized directly to MXFP4 to leverage highly optimized hardware kernels.

5. Bonus Single-Quant (MXFP4)

• Single Tier (All Layers - MXFP4 Precision): All layers are target-quantized directly to MXFP4, for speed and a low VRAM footprint.

📝 Exact Conversion Details

Generate Tri-Quant MXFP4_MOE + Q8_0 + F16:

Generate Dual-Quant MXFP4_MOE + Q8_0:

Generate Tri-Quant MXFP4_MOE + Q6_K + Q8_0:

Generate Dual-Quant MXFP4_MOE + Q6_K:

Generate Single-Quant MXFP4_MOE:

ℹ️ Misc Details

🤝 Support the Journey

✨ Acknowledgments

• LatitudeGames for the exceptional Equinox-31B base model.
• bartowski for the BF16 GGUF quants of Equinox-31B.
• mradermacher for the high-quality importance matrix data anchors.