license: apache-2.0
language:

• en
• zh
  pipeline_tag: text-generation
  tags:
• qwen
• qwen3.8
• dense
• efficient-thinking
• thinking
• reasoning
• coding
• agent
• conversational
• uncensored
• heretic
• sft
• simpo
• bf16
• fp8
• block128
• dflash2
• speculative-decoding
• sglang
• vllm

Qwen3.8-27B EfficientThink · SFT → SimPO · DFlash2

Refusal evaluation: 4 / 140 (2.9%) across 10 categories and 140 prompts.

Research disclaimer: This experimental release is provided solely to study the technical feasibility and behavioral effects of refusal-tendency dissolution. It is not a comprehensive safety conclusion, an endorsement of unrestricted use, or professional advice. Users are responsible for lawful and appropriate use and for independently verifying model outputs.

Static FP8 + DFlash2 measured serving results

Measured on DGX Spark with the published static Block128 FP8 main model, SGLang + DFlash2, XH, draft_tokens=8, and max_tokens=256. All six tiers completed without request errors. C24 maximizes aggregate throughput; C1 is the lowest-concurrency/highest-accepted-length tier; C8 is the recommended practical balance. finish_reason=length is expected in this fixed-length pressure test and is not a quality judgment.

A separate C3, max_tokens=1024 science/code/general smoke passed 3/3, returned 3/3 stop, had no empty output or mojibake, and delivered 68 tok/s aggregate.

Tested launcher (pass downloaded paths explicitly):

bash FP8/runtime/TESTED_STARTUP_C24_DFLASH2.sh \
  "$PWD/FP8" "$PWD/FP8/DFlash2-FP8" efficientthink-fp8-dflash2 19110

EfficientThink targets unproductive reasoning tails—not reasoning itself. It is trained to preserve capability and genuinely necessary long reasoning while improving terminal-answer reliability.

This repository contains the final merged SimPO BF16 model under BF16/ and a text-only static Block128 FP8 main model under FP8/. For one-directory downloads, both model directories include the same verified optional DFlash2 draft under their own DFlash2-FP8/ subdirectory; the draft does not replace the main model.

Artifacts

Download either BF16/ or FP8/ to receive the corresponding main model and its DFlash2 runtime files together. FP8/manifest.json and FP8/SHA256SUMS define the static-FP8 main package. The FP8 main model is language-only: 64 layers, 1,251 tensors after SGLang key repack, zero visual tensors, and zero MTP tensors.

Final SimPO evaluation

Frozen XH protocol, max_tokens=32768; every non-passing sample remains in the denominator.

LiveCodeBench breakdown: easy 23/23, medium 27/31, hard 24/46. The 74/100 score is the final all-failures-counted operational result; it is not labeled as a clean run.

Same-protocol capability and reasoning comparison

Protocol: dynamic FP8 + DFlash2, 2× GPU C24, XH, max_tokens=32768; official full-suite results. All failures remain in the denominator.

GPQA Diamond · 198 questions

GPQA improves by 3.54pp while mean reasoning, truncation, unparseable outputs, and loose loop candidates fall.

MMLU · 500 questions

MMLU reasoning cost and long-tail incidence fall, but accuracy also drops by 1.80pp. This is reported as a real capability trade-off, not hidden behind the efficiency gain.

LiveCodeBench · 100 questions

Reasoning statistics below cover all 100 cases; timeout/error rows contribute zero reasoning tokens.

LCB gains are concentrated in hard problems and submission reliability. The run does not show an overall shortening of code reasoning: mean, P50, and >8K counts rise slightly, while >16K and 32K truncations remain unchanged. SimPO converts some former non-submissions into valid solutions without eliminating the 32K tail.

Training

Qwen/Qwen3.8-27B → capability-preserving SFT → terminal-behavior SimPO → per-tensor FP32 delta merge → BF16.

SFT · 1,905 examples

• 1 epoch · 239 optimizer steps · effective batch 8
• LoRA r=16 · alpha=32 · dropout=0.05
• LR 5e-6 · 12 warmup steps · seed 20260901
• 2× NVIDIA RTX PRO 6000 Blackwell Server Edition

SimPO · 110 preference pairs / 73 unique prompts

• 5 optimizer steps · beta=1.0 · gamma=0.2 · peak LR 5e-7
• LoRA r=16 · alpha=32 · dropout=0
• seed 20260903 · world size 2 · FSDP full sharding

Recommended Transformers usage

from transformers import AutoModelForCausalLM, AutoTokenizer

repo = "nerkyor/Qwen3.8-27B-EfficientThink-Uncensored-K3-Opus5-Grok4.6-GPT5.6Sol-SFT-SimPO-DFlash2"
tok = AutoTokenizer.from_pretrained(repo, subfolder="BF16", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    repo,
    subfolder="BF16",
    torch_dtype="auto",
    device_map="auto",
    trust_remote_code=True,
)

messages = [{"role": "user", "content": "Explain your answer clearly."}]
prompt = tok.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True,
    enable_thinking=True,
    reasoning_effort="xhigh",
)
inputs = tok(prompt, return_tensors="pt").to(model.device)
output = model.generate(
    **inputs,
    max_new_tokens=4096,
    do_sample=True,
    temperature=1.0,
    top_p=0.95,
    top_k=20,
)
print(tok.decode(output[0][inputs.input_ids.shape[-1]:], skip_special_tokens=True))

The bundled generation defaults are temperature=1.0, top_p=0.95, and top_k=20. Set enable_thinking=False for non-thinking mode. The frozen capability and serving results reported here use reasoning_effort="xhigh".

DFlash2 scope and limits

• The same verified draft payload is bundled at BF16/DFlash2-FP8/ and FP8/DFlash2-FP8/, each with model.safetensors, config.json, manifest.json, and SHA256SUMS.
• The frozen evaluation configuration used DFLASH, one speculative step, top-k 1, eight draft tokens, block size 8, Triton draft attention, and an FP8 draft.
• Runtime support for this draft must be verified against the serving stack in use. The draft is not a standalone chat model.
• Scores are tied to the frozen XH harness and serving configuration and should not be compared across unrelated harnesses.
• Long reasoning can still be necessary; EfficientThink is not a universal short-answer mode.
• Verify outputs independently, especially in high-stakes contexts.

Recommended deployment: SGLang

Use SGLang for this release. It is the runtime used for the frozen FP8+DFlash2 quality, acceptance, and concurrency measurements. For XH thinking mode, use Qwen's official sampling preset: temperature=1.0, top_p=0.95, top_k=20, min_p=0.0, presence_penalty=0.0, and repetition_penalty=1.0.

SGLang — verified FP8 + DFlash2 path

bash FP8/runtime/TESTED_STARTUP_C24_DFLASH2.sh \
  "$PWD/FP8" "$PWD/FP8/DFlash2-FP8" efficientthink-fp8-dflash2 19110

Use C8 for the measured practical balance or C24 when maximum aggregate throughput is the priority. This is the repository's frozen, tested DFlash2 path.

vLLM — official baseline reference only, not tested here

vllm serve "$PWD/FP8" \
  --served-model-name efficientthink-fp8

Pass reasoning_effort="xhigh" and the official thinking sampling preset in the OpenAI-compatible request. This command is included only as the official Qwen3.8 baseline serving shape. This repository has not tested vLLM for this checkpoint and does not recommend or claim a frozen vLLM+DFlash2 result.

References: official Qwen3.8-27B model card, SGLang documentation, and vLLM documentation.

中文说明

拒答评测：10 个类别、140 条提示中为 4 / 140（2.9%）。

科研免责声明：本实验版本仅用于研究拒答倾向消解的技术可行性及其行为影响。它不构成全面的安全结论，不代表对无限制使用的认可，也不构成任何专业建议。用户应依法、恰当地使用，并独立核验模型输出。

静态 FP8 + DFlash2 实测服务结果

实测环境为 DGX Spark、已发布的静态 Block128 FP8 主模型、SGLang + DFlash2、XH、draft_tokens=8、max_tokens=256。六档均无请求错误。C24 为最大聚合吞吐；C1 为最低并发且平均接受长度最高；日常实用平衡推荐 C8。定长压力测试中的 finish_reason=length 是测试设计，不作为质量判断。

另行执行的 C3、max_tokens=1024 science/code/general 质量 smoke 为 3/3 正确、3/3 stop、无空答/乱码、聚合 68 tok/s。

实测启动脚本（显式传入下载目录）：

bash FP8/runtime/TESTED_STARTUP_C24_DFLASH2.sh \
  "$PWD/FP8" "$PWD/FP8/DFlash2-FP8" efficientthink-fp8-dflash2 19110

EfficientThink 优化的是无效推理长尾，而不是推理本身。 目标是在保留能力和必要长推理的同时，提高最终答案的可靠性。

本仓包含 BF16/ 下的最终合并 SimPO BF16 主模型，以及 FP8/ 下的纯文本静态 Block128 FP8 主模型。为便于按目录一次下载，两个主模型目录内均配套同一份已验证的可选 DFlash2-FP8/ draft；draft 不替代主模型。

文件

下载 BF16/ 或 FP8/ 任一目录，即可同时取得对应主模型与 DFlash2 运行文件。静态 FP8 主包以 FP8/manifest.json 与 FP8/SHA256SUMS 为准；该 FP8 主模型为纯文本：64 层、SGLang 键名重排后 1,251 tensors、visual tensors=0、MTP tensors=0。

最终 SimPO 冻结评测

冻结 XH 协议，max_tokens=32768；全部未通过样本均保留在分母中。

LiveCodeBench 难度分布：easy 23/23、medium 27/31、hard 24/46。74/100 是将全部失败计入后的最终 operational 分数，不标注为 clean run。

同协议能力与思考对比

协议：动态 FP8 + DFlash2、双卡 C24、XH、max_tokens=32768，原版与最终 SimPO 均为正式全量结果；所有失败均保留在分母中。

GPQA Diamond｜198题

GPQA 提升 3.54pp，同时平均思考、截断、不可解析与宽松 LOOP 候选均下降。

MMLU｜500题

MMLU 的平均思考与长尾明显下降，但正确率同步回落 1.80pp；这是需要如实披露的能力交换，不能只展示效率改善。

LiveCodeBench｜100题

下列 reasoning 统计按全部 100 题计算，超时/错误题以 0 reasoning tokens 计入。

LCB 的增益主要来自难题与提交可靠性。代码推理没有整体缩短：平均、P50 和 >8K 略增，>16K 与 32K 截断不变。也就是说，SimPO 将一部分原本无法提交的样本转化为有效解答，但尚未进一步消除 32K 长尾。

训练

Qwen/Qwen3.8-27B → 能力保持 SFT → 终止行为 SimPO → 逐 tensor FP32 delta 合并 → BF16。

SFT · 1,905 条样本

• 1 epoch · 239 optimizer steps · effective batch 8
• LoRA r=16 · alpha=32 · dropout=0.05
• LR 5e-6 · warmup 12 steps · seed 20260901
• 双 NVIDIA RTX PRO 6000 Blackwell Server Edition

SimPO · 110 组偏好对 / 73 个唯一 prompt

• 5 optimizer steps · beta=1.0 · gamma=0.2 · peak LR 5e-7
• LoRA r=16 · alpha=32 · dropout=0
• seed 20260903 · world size 2 · FSDP full sharding

推荐 Transformers 用法

from transformers import AutoModelForCausalLM, AutoTokenizer

repo = "nerkyor/Qwen3.8-27B-EfficientThink-Uncensored-K3-Opus5-Grok4.6-GPT5.6Sol-SFT-SimPO-DFlash2"
tok = AutoTokenizer.from_pretrained(repo, subfolder="BF16", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    repo, subfolder="BF16", torch_dtype="auto", device_map="auto", trust_remote_code=True
)

messages = [{"role": "user", "content": "请清晰说明你的结论。"}]
prompt = tok.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True,
    enable_thinking=True,
    reasoning_effort="xhigh",
)
inputs = tok(prompt, return_tensors="pt").to(model.device)
output = model.generate(
    **inputs,
    max_new_tokens=4096,
    do_sample=True,
    temperature=1.0,
    top_p=0.95,
    top_k=20,
)
print(tok.decode(output[0][inputs.input_ids.shape[-1]:], skip_special_tokens=True))

模型内置生成默认值为 temperature=1.0、top_p=0.95、top_k=20。非思考模式可设置 enable_thinking=False；本卡披露的冻结能力与服务结果均使用 reasoning_effort="xhigh"。

DFlash2 范围与限制

• 同一份已验证 draft 分别随包放在 BF16/DFlash2-FP8/ 与 FP8/DFlash2-FP8/，均包含 model.safetensors、config.json、manifest.json 与 SHA256SUMS。
• 冻结评测配置使用 DFLASH、1 speculative step、top-k 1、8 draft tokens、block size 8、Triton draft attention 和 FP8 draft。
• 使用前必须在目标服务框架中核对兼容性；draft 不能作为独立聊天模型运行。
• 分数绑定冻结 XH harness 与服务配置，不用于跨协议直接比较。
• 必要的长推理仍然保留；EfficientThink 不是统一短答模式。
• 高风险场景请独立核验输出。

推荐部署：SGLang

本模型推荐使用 SGLang。 冻结的 FP8+DFlash2 质量、接受率与并发数据均由 SGLang 实测获得。XH 思考模式采用 Qwen 官方采样参数：temperature=1.0、top_p=0.95、top_k=20、min_p=0.0、presence_penalty=0.0、repetition_penalty=1.0。

SGLang — 已验证 FP8 + DFlash2 路径

bash FP8/runtime/TESTED_STARTUP_C24_DFLASH2.sh \
  "$PWD/FP8" "$PWD/FP8/DFlash2-FP8" efficientthink-fp8-dflash2 19110

日常部署推荐采用实测平衡档 C8；追求最大聚合吞吐时使用 C24。这是本仓已冻结实测的 DFlash2 路径。

vLLM — 仅列官方基础参考，本仓未实测

vllm serve "$PWD/FP8" \
  --served-model-name efficientthink-fp8

OpenAI-compatible 请求中显式传入 reasoning_effort="xhigh" 与官方思考采样参数。该命令仅作为 Qwen3.8 官方基础服务形态参考；本仓没有实测该 checkpoint 的 vLLM 路径，不推荐也不声称已完成 vLLM+DFlash2 冻结实测。

参考：Qwen3.8-27B 官方模型卡、SGLang 文档 与 vLLM 文档。