What are the hardware requirements for Parm 2 CoT 14B 16K O1 QwQ?

Parm 2 CoT 14B 16K O1 QwQ requires approximately 29.4 GB of VRAM and supports a context window of 16K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Parm 2 CoT 14B 16K O1 QwQ and how large is it?

Parm 2 CoT 14B 16K O1 QwQ is developed by Pinkstack, a model with 14b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Parm 2 CoT 14B 16K O1 QwQ perform on standard benchmarks?

Parm 2 CoT 14B 16K O1 QwQ has the following published scores: LMSYS ELO 1203. Compare against reference models on this page or on the LLM Explorer leaderboards.

Parm 2 CoT 14B 16K O1 QwQ by Pinkstack — VRAM 29.4GB, 16K context

Name: Parm 2 CoT 14B 16K O1 QwQ
Rating: 2.38 (8 reviews)
Author: Pinkstack

Parm 2 CoT 14B 16K O1 QwQ is an open-source language model by Pinkstack. Features: 14b LLM, VRAM: 29.4GB, Context: 16K, License: mit, Quantized, LLM Explorer Score: 0.28, ELO: 1203.

Arxiv:2412.08905 4bit Base model:finetune:microsoft/... Base model:microsoft/phi-4 Chat Code Conversational Cot Dataset:amphora/qwq-longcot-13... En Endpoints compatible Llama Math Model-index Multilingual Phi Phi3 Pytorch Quantized Reasoning Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/Pinkstack/Parm-2-CoT-14B-16k-o1-QwQ

Parm 2 CoT 14B 16K O1 QwQ Benchmarks

LMSys ELO 1203 vs 1272 (so35) -5.4%

^nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

What is the LLM Explorer Rank (Score)

Parm 2 CoT 14B 16K O1 QwQ Parameters and Internals

LLM Name	Parm 2 CoT 14B 16K O1 QwQ
Repository 🤗	https://huggingface.co/Pinkstack/Parm-2-CoT-14B-16k-o1-QwQ
Base Model(s)	Phi 4 microsoft/phi-4
Model Size	14b
Required VRAM	29.4 GB
Updated	2026-08-04
Maintainer	Pinkstack
Model Type	llama
Model Files	4.9 GB: 1-of-6 5.0 GB: 2-of-6 4.9 GB: 3-of-6 5.0 GB: 4-of-6 5.0 GB: 5-of-6 4.6 GB: 6-of-6
Supported Languages	en
Quantization Type	4bit
Model Architecture	LlamaForCausalLM
License	mit
Context Length	16384
Model Max Length	16384
Transformers Version	4.47.1
Tokenizer Class	GPT2Tokenizer
Padding Token	<\|dummy_87\|>
Vocabulary Size	100352
Torch Data Type	float16

Rank the Parm 2 CoT 14B 16K O1 QwQ capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to Parm 2 CoT 14B 16K O1 QwQ

Best Alternatives	Context / RAM	Downloads	Likes
Phi 4 14B Grpo Gsm8k 3e	16K / 29.4 GB	8	0
Phi 4 14B Grpo Limo	16K / 29.4 GB	7	0
...perThoughts CoT 14B 16K O1 QwQ	16K / 29.4 GB	265	6
Heliars Phi4 Carla X1 14B	16K / 29.4 GB	3	2
...hts CoT 14B 16K O1 QwQ PyTorch	16K / 29.4 GB	0	1
CausalLM 14B EXL2	8K / 8.5 GB	8	3
...struct 2512 Llamafied TextOnly	256K / 27 GB	9	3
...Qwen2.5llamaify 14B V23.3 200K	195K / 29.7 GB	10	5
...Qwen2.5llamaify 14B V23.1 200K	195K / 29.7 GB	15	1
GeM2 Llamion 14B LongChat	195K / 29 GB	7915	4