What are the hardware requirements for MAmmoTH Coder 34B AWQ?

MAmmoTH Coder 34B AWQ requires approximately 18.3 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 MAmmoTH Coder 34B AWQ and how large is it?

MAmmoTH Coder 34B AWQ is developed by TheBloke, a model with 34b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate MAmmoTH Coder 34B AWQ?

MAmmoTH Coder 34B AWQ is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

MAmmoTH Coder 34B AWQ by TheBloke — VRAM 18.3GB, 16K context

Name: MAmmoTH Coder 34B AWQ
Author: TheBloke

MAmmoTH Coder 34B AWQ is an open-source language model by TheBloke. Features: 34b LLM, VRAM: 18.3GB, Context: 16K, License: mit, Quantized, Code Generating, LLM Explorer Score: 0.09.

Arxiv:2309.05653 4-bit Awq Base model:quantized:tiger-lab... Base model:tiger-lab/mammoth-c... Codegen Dataset:tiger-lab/mathinstruct En Llama Quantized Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/TheBloke/MAmmoTH-Coder-34B-AWQ

MAmmoTH Coder 34B AWQ Parameters and Internals

Model Type

llama

Use Cases

Areas:

research, educational software, tutoring systems

Primary Use Cases:

Solving general math problems

Limitations:

Performance may vary based on complexity and specifics of the math problem, Not all mathematical fields comprehensively covered

Additional Notes

Quantized by TheBloke using AWQ method for compact model size and efficient inference.

Training Details

Data Sources:

MathInstruct Dataset

Methodology:

Hybrid Instruction Tuning using chain-of-thought and program-of-thought rationales.

Input Output

Input Format:

Instruction-based prompt

Output Format:

Response

LLM Name	MAmmoTH Coder 34B AWQ
Repository 🤗	https://huggingface.co/TheBloke/MAmmoTH-Coder-34B-AWQ
Model Name	MAmmoTH Coder 34B
Model Creator	TIGER-Lab
Base Model(s)	MAmmoTH Coder 34B TIGER-Lab/MAmmoTH-Coder-34B
Model Size	34b
Required VRAM	18.3 GB
Updated	2026-07-24
Maintainer	TheBloke
Model Type	llama
Model Files	9.9 GB: 1-of-2 8.4 GB: 2-of-2
Supported Languages	en
AWQ Quantization	Yes
Quantization Type	awq
Generates Code	Yes
Model Architecture	LlamaForCausalLM
License	mit
Context Length	16384
Model Max Length	16384
Transformers Version	4.31.0
Tokenizer Class	LlamaTokenizer
Beginning of Sentence Token	<s>
End of Sentence Token	</s>
Unk Token	<unk>
Vocabulary Size	32001
Torch Data Type	bfloat16

Rank the MAmmoTH Coder 34B AWQ 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 MAmmoTH Coder 34B AWQ

Best Alternatives	Context / RAM	Downloads	Likes
...echless Codellama 34B V2.0 AWQ	16K / 18.3 GB	6	6
Phind CodeLlama 34B V2 AWQ	16K / 18.3 GB	42	35
...enbuddy Coder 34B V11 Bf16 AWQ	16K / 18.5 GB	4	1
CodeLlama 34B Instruct AWQ	16K / 18.3 GB	57	1
...izardCoder Python 34B V1.0 AWQ	16K / 18.3 GB	11	6
CodeLlama 34B AWQ	16K / 18.3 GB	24	2
CodeFuse CodeLlama 34B AWQ	16K / 18.3 GB	11	2
CodeLlama 34B Python AWQ	16K / 18.3 GB	8	2
...nd CodeLlama 34B Python V1 AWQ	16K / 18.3 GB	7	2
Phind CodeLlama 34B V1 AWQ	16K / 18.3 GB	6	1

Note: green Score (e.g. "73.2") means that the model is better than TheBloke/MAmmoTH-Coder-34B-AWQ.