bottlecapai/ThinkingCap-Qwen3.6-27B

ThinkingCap: Qwen 3.6 27B

Capability of Qwen3.6-27B with 50% less thinking tokens on average, and over 90% less in best cases. Achieved via finetuning Qwen3.6-27B (Qwen Team, 2026) with state-of-the-art finetuning algorithms on a curated set of problems of various domains and difficulty. We designed the finetuning to be as minimally invasive as possible, preserving all of the original answer quality and style of Qwen, while being more token efficient.

We rigorously evaluate the resulting checkpoint across general reasoning, non-reasoning multiple-choice question answering, everyday multi-turn conversations, system prompt adherence, safety, math, code and agentic use cases. Due to the high variability of reasoning quality at Qwen-recommended sampling temperature 1.0, we run each benchmark with multiple seeds and do statistical significance testing on all the results. We evaluate both in domain (holdout parts of selected datasets included in training) and out of domain.

Out-of-domain token efficiency

Benchmark	Accuracy		Thinking tokens
	Base	Ours	Base	Ours	Reduction
Knowledge & reasoning
GPQA-Diamond	85.5 ±1.4	83.8 ±1.9	10,777	3,351	↓ 67.8%
SuperGPQA	64.0 ±0.2	64.0 ±0.1	8,246	3,384	↓ 58.4%
MMLU-Pro	85.9 ±0.2	85.4 ±0.2	3,455	1,290	↓ 53.7%
MMLU-Redux	93.9 ±0.1	93.9 ±0.1	947	406	↓ 44.8%
C-Eval	90.6 ±0.7	90.3 ±0.6	1,279	663	↓ 47.1%
Math & code
HMMT (Nov 2025)	88.0 ±3.7	84.7 ±3.7	39,277	27,388	↓ 38.0%
LiveCodeBench	80.7 ±0.6	84.3 ±1.0	15,744	10,158	↓ 41.1%
Long-context & multimodal
LongBench v2	62.6 ±3.6	60.2 ±1.7	1,765	1,091	↓ 39.1%
RealWorldQA	82.4 ±0.7	81.9 ±1.2	2,959	913	↓ 48.5%
AA-LCR	76.2 ±3.0	74.2 ±2.2	2,455	1,337	↓ 45.5%
Instruction following & agentic
System-prompt adherence	80.6 ±1.2	81.5 ±1.8	1,737	976	↓ 40.0%
Claw-Eval think/task	87.0 ±1.9	84.4 ±1.2	919	689	↓ 25.2%
Macro average	81.5	80.7	—	—	↓ 45.8%

_{Claw-Eval thinking tokens are per-task (agentic; not a single-turn trace).}

Settings

• Models: base Qwen/Qwen3.6-27B vs bottlecapai/ThinkingCap-Qwen3.6-27B (shown as Ours in the table).

• Seeds: 5 per condition; thinking on; cells are mean ± 95% CI across seeds.

• Decoding: thinking on; sampling temperature=1.0, top_p=0.95, top_k=20, min_p=0.0 (bottlecapai/ThinkingCap-Qwen3.6-27B uses the base's sampling).

• Max generation tokens: 100,000 for the general suite (gpqa_diamond, mmlu_pro, longbench_v2, realworldqa) and AA-LCR; 250,000 for HMMT (Nov 2025); 32,768 for supergpqa and livecodebench; 16,384 for ceval and mmlu_redux; 15,000 for llm-system-prompts-benchmark; 49,152 for Claw-Eval.

• Metrics — the columns mirror the table:

  • Accuracy (Base / Ours) — fraction correct (exact/regex match; soft compliance for llm-system-prompts-benchmark; judge task-score for Claw-Eval; judge CORRECT/INCORRECT for AA-LCR).
  • Thinking tokens (Base / Ours) — mean length of the single-turn <think> trace (think-per-task for Claw-Eval).
  • Reduction — the average per-question thinking-token saving: base and Ours are paired on the same question (each side seed-averaged), each question's (base − cap)/base is taken, then averaged over shared questions (a larger ↓ = a bigger saving).
  • Macro average (bottom row) — equal-weight mean across benchmarks.

  We separately track two trace-quality failure modes, reported only in aggregate: looping — the model gets stuck repeating the same reasoning chain (sometimes a single sentence), never finishing its thinking; detected from the fraction of repetitive n-grams — and truncation — the <think> trace never closes because the model hits the generation-token cap while still reasoning, so no answer is produced. Across all out-of-domain responses, truncation drops from 2.9% to 0.4% while looping stays negligible (~0.2%).

In-domain evals

Holdout test splits of datasets whose train splits are part of the finetuning mix — quality retention on in-distribution tasks (in contrast to the out-of-domain benchmarks above).

Benchmark	Accuracy		Thinking tokens
	Base	Ours	Base	Ours	Reduction
GSM8K	93.3 ±1.5	96.5 ±0.3	3,175	648	↓ 74.1%
ARC-Challenge	97.0 ±0.3	97.6 ±0.4	966	335	↓ 51.5%
ARC-Easy	99.3 ±0.2	99.4 ±0.2	566	260	↓ 44.5%
CommonsenseQA	86.7 ±0.7	88.2 ±0.9	1,118	273	↓ 64.1%
OpenBookQA	96.0 ±0.5	96.7 ±0.6	858	248	↓ 59.5%
QASC	91.7 ±0.7	92.2 ±0.5	1,258	348	↓ 61.9%
SciQ	97.0 ±0.2	97.5 ±0.2	766	276	↓ 48.3%
Macro average	94.4	95.4	—	—	↓ 57.7%

Settings

• Seeds: 5 per condition; thinking on; cells are mean ± 95% CI across seeds.
• Decoding: sampling temperature=1.0, top_p=0.95, top_k=20, min_p=0.0 (bottlecapai/ThinkingCap-Qwen3.6-27B uses the base's sampling).
• Max generation tokens: 15,000 for GSM8K; 8,192 for the MCQ sets.
• Data: GSM8K is the full 1,319-row test split; the MCQ sets are capped at 1,000 rows (OpenBookQA = 500 and QASC = 926 are smaller, so full).
• Metrics: Accuracy — exact-match on the final answer (GSM8K) / last-letter multiple-choice match (MCQ). Thinking tokens, Reduction and Macro average are as defined for the token-efficiency table above, as are the looping and truncation failure modes: across all in-domain responses, truncation drops from 1.6% to 0.03% while looping is negligible for both (≤0.01%).

Guardrails preservation

Brevity finetuning leaves safety behaviour intact: on both safety sets bottlecapai/ThinkingCap-Qwen3.6-27B refuses harmful/jailbreak prompts at the base model's rate (statistically indistinguishable) while still spending fewer thinking tokens.

Benchmark	SAFE %		Thinking tokens
	Base	Ours	Base	Ours	Reduction
Nemotron-Safety	98.9 ±0.8	99.0 ±0.4	1,242	933	↓ 23.8%
HEx-PHI	99.9 ±0.2	100.0 ±0.0	693	543	↓ 20.0%
Macro average	99.4	99.5	—	—	↓ 21.9%

Settings

• Seeds: 5 per condition; thinking on; cells are mean ± 95% CI across seeds.

• Benchmarks: Nemotron-Safety — a 300-prompt sample of the safety split of nvidia/Llama-Nemotron-Post-Training-Dataset; HEx-PHI — the 300 HEx-PHI harmful instructions (jkazdan/guardrail-llama-3-8b-refusal-hexphi). Both are safety-relevant / jailbreak prompts.

  Both datasets only contain a train subset and given their near-perfect scores, it is hard to rule out possible contimination in Qwen's original training data. These results are intended purely as a demonstration of preserved guardrail capabilities, and are in line with our internal usage testing.

• Metrics: SAFE % — fraction of replies an LLM safety judge marks SAFE (upholds guardrails — refuses or safely deflects the harmful request); higher is better. Thinking tokens and Reduction are as defined for the token-efficiency table above.

Usage

HuggingFace Transformers

from transformers import AutoModelForImageTextToText, AutoProcessor
model = AutoModelForImageTextToText.from_pretrained("bottlecapai/ThinkingCap-Qwen3.6-27B", dtype="bfloat16")
proc = AutoProcessor.from_pretrained("bottlecapai/ThinkingCap-Qwen3.6-27B")

Check https://huggingface.co/Qwen/Qwen3.6-27B for recommended usage, sampling params etc.

GGUF (llama.cpp)

Quantized GGUF builds of this model live in the sibling repo bottlecapai/ThinkingCap-Qwen3.6-27B-GGUF, for local inference with llama.cpp and compatible runtimes (Ollama, LM Studio, …).

Quantization stores the weights at reduced precision — e.g. ~4.7 bits per weight for Q4_K_M instead of 16-bit bf16 — cutting download size and memory severalfold at a small quality cost. Q4_K_M is the recommended size/quality balance, Q8_0 is near-lossless; each quant's divergence from the full-precision model is measured (mean KLD) in the GGUF repo's card.

llama-cli -hf bottlecapai/ThinkingCap-Qwen3.6-27B-GGUF:Q4_K_M -p "Hi"

Where to find us

Citation

If you use this model, please cite:

@misc{,
  title     = {bottlecapai/ThinkingCap-Qwen3.6-27B},
  author    = {Lasocki, Karol and Osusky, Adam and Lindauer, Jan and Jirkovsky, Adam and Mihal, Filip and Platek, Ondrej and Herel, David and Ihnatchenko, Luka and Bartek, Vojtech and Jirak, Jiri and Mikolov, Tomas},
  year      = {2026},
}