SAVRN
Search Contact SAVRN

Open-weight model

sami-27b-mixbit

by Tchbcn tchbcb/sami-27b-mixbit

samai-27b(Qwen/Qwen3.8-27B 手术版:PonderNet 停机位点 L35/39/43/47/51/55/59/63 + dMoE 8 专家 + MTP)经 QLoRA SFT 后,按 mixbit 静态化管线产出的 llama.cpp 混合精度权重。 等价稠密模型 ≙ Qwen35ForCausalLM(48 linearattention + 16 fullattention,hidden 5120),可直接用上游 llama.cpp 载入。 - Token embedding /…

Parameters
Context
Weights34.6 GB
Licenseapache-2.0
AccessOpen weights
Monthly Downloads

Model Card

By Tchbcn, published under apache-2.0, revision c39920a07eb2.

samai-27b(Qwen/Qwen3.8-27B 手术版:PonderNet 停机位点 L35/39/43/47/51/55/59/63 + dMoE 8 专家 + MTP)经 QLoRA SFT 后,按 mixbit 静态化管线产出的 llama.cpp 混合精度权重。 等价稠密模型 ≙ Qwen35ForCausalLM(48 linearattention + 16 fullattention,hidden 5120),可直接用上游 llama.cpp 载入。 - Token embedding / output:Q6K(--token-embedding-type Q6K) OpenAI 兼容端点 /v1/chat/completions。 - 基座:tchbcb/samai-27b(含 SFT adapter / 训练日志 / 数据) - 配方参考:tchbcb/samai-4b-mixbit

Read Tchbcn's full model card

samai-27b-mixbit (Q5_K_M mixed-precision GGUF)

samai-27b(Qwen/Qwen3.8-27B 手术版:PonderNet 停机位点 L35/39/43/47/51/55/59/63 + dMoE 8 专家 + MTP)经 QLoRA SFT 后,按 mixbit 静态化管线产出的 llama.cpp 混合精度权重。

静态化(相对 samai-27b 的结构变化)

组件 处理
PonderNet (L35..63 停机循环) 展开:ponder_forced_steps=1 单遍语义,halter/router 丢弃
dMoE 8 专家 (仅 L35..63) 按校准激活频率 f_i 加权合并为稠密 FFN(8×2176 → 17408,与稠密层同形)
MTP 模块 丢弃(推理旁路;训练时 mtp_loss_weight=0)
视觉塔 (460M) 丢弃(纯文本)
LoRA SFT (r16 α32, 124.66M) 已合并进基座权重

等价稠密模型 ≙ Qwen3_5ForCausalLM(48 linear_attention + 16 full_attention,hidden 5120),可直接用上游 llama.cpp 载入。

量化配方(与 samai-4b-mixbit 同款)

  • 主体:Q5_K_M
  • Token embedding / output:Q6_K--token-embedding-type Q6_K

文件

文件 说明
samai-27b-mixbit-Q5_K_M.gguf 主权重(含 Q6_K embedding)
sft-adapter/(见 samai-27b 仓) SFT 训练侧信息

用法

llama-server -m samai-27b-mixbit-Q5_K_M.gguf --host 0.0.0.0 --port 8100 -ngl 999 -c 131072 -fa

OpenAI 兼容端点 /v1/chat/completions

来源

  • 基座:tchbcb/samai-27b(含 SFT adapter / 训练日志 / 数据)
  • 配方参考:tchbcb/samai-4b-mixbit

Identity and Version

Repository
tchbcb/sami-27b-mixbit
Publisher
Tchbcn
Task
Not stated by the source
Modality
Other
Library
gguf
Parameters
Not stated by the source
Languages
Not stated by the source
Revision
c39920a07eb2e9f0daaa1635a580759bfb93b57f
First published
2026-09-17
Last updated
2026-09-18

Files and Weights

15 files, 34.6 GB in total. The weights are 2 files totalling 34.6 GB in gguf.

Weights2 files · 34.6 GB
Documentation1 file · 1.7 KB
Other11 files · 403.5 KB
Repository1 file · 1.6 KB
Every file
FileTypeSizeSHA-256
samai-27b-mixbit-IQ4_XS.ggufWeights15.2 GB 05f2da870c00
samai-27b-mixbit-Q5_K_M.ggufWeights19.4 GB 9ea172a5c9dc
README.mdDocumentation1.7 KB
logs/0918_001830_master8.logOther443 B
logs/0918_001830_pip_core.logOther637 B
logs/0918_001830_preflight.logOther164 B
logs/0918_001830_preflight_help.logOther6.0 KB
logs/0918_005127_convert.logOther128.4 KB
logs/0918_005128_master8.logOther875 B
logs/0918_005128_quant_i4.logOther4.9 KB
logs/0918_010111_convert.logOther128.4 KB
logs/0918_023427_master8.logOther1.8 KB
logs/0918_023427_push.logOther3.5 KB
logs/0918_024743_convert.logOther128.4 KB
.gitattributesRepository1.6 KB

License and Download

License
apache-2.0
Access
Open weights, no gate
Download size
34.6 GB
Download from Tchbcn

Released by Tchbcn through its official repository on Hugging Face. Read the license.

Built From

  • Derived from tchbcb/samai-27b
  • Quantized from tchbcb/samai-27b

Memory Requirements

PrecisionWeights in memory
As published34.6 GB

Weights only, from the published parameter count; the key-value cache and runtime add to this.

Questions About sami-27b-mixbit

Can I use sami-27b-mixbit commercially?

Yes. sami-27b-mixbit is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.