What are the hardware requirements for Typhoon 7B?

Typhoon 7B requires approximately 14.6 GB of VRAM and supports a context window of 32K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Typhoon 7B and how large is it?

Typhoon 7B is developed by scb10x, a model with 7b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Typhoon 7B perform on standard benchmarks?

Typhoon 7B has the following published scores: MMLU 59.54. Compare against reference models on this page or on the LLM Explorer leaderboards.

Typhoon 7B by scb10x — VRAM 14.6GB, 32K context

Name: Typhoon 7B
Author: scb10x

Typhoon 7B is an open-source language model by scb10x. Features: 7b LLM, VRAM: 14.6GB, Context: 32K, License: apache-2.0, LLM Explorer Score: 0.13, Arc: 58.5, HellaSwag: 81.6, MMLU: 59.5, GSM8K: 31.6.

Arxiv:2312.13951 Autotrain compatible Endpoints compatible Mistral Pretrained Pytorch Region:us Safetensors Sharded Tensorflow Th

Model Card on HF 🤗: https://huggingface.co/scb10x/typhoon-7b

Typhoon 7B Parameters and Internals

Model Type

pretrained, decoder-only, text-generation

Use Cases

Limitations:

May not follow human instructions without one/few-shot learning or instruction fine-tuning, Does not have moderation mechanisms and may generate harmful or inappropriate responses

Additional Notes

The model does not have any moderation mechanisms, and may generate harmful or inappropriate responses. Follow the development on Twitter or join the Discord community for support.

Supported Languages

Thai (on par with GPT-3.5), English (primary language)

LLM Name	Typhoon 7B
Repository 🤗	https://huggingface.co/scb10x/typhoon-7b
Model Size	7b
Required VRAM	14.6 GB
Updated	2025-11-03
Maintainer	scb10x
Model Type	mistral
Model Files	10.0 GB: 1-of-2 4.6 GB: 2-of-2 10.0 GB: 1-of-2 4.6 GB: 2-of-2
Supported Languages	th
Model Architecture	MistralForCausalLM
License	apache-2.0
Context Length	32768
Model Max Length	32768
Transformers Version	4.34.0.dev0
Tokenizer Class	LlamaTokenizer
Vocabulary Size	35219
Torch Data Type	bfloat16

Rank the Typhoon 7B 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

Quantized Models of the Typhoon 7B

Model	Likes	Downloads	VRAM
Typhoon 7B GGUF	9	1555	3 GB
Typhoon 7B GGUF	1	226	2 GB
Typhoon 7B GPTQ	4	6	4 GB
Typhoon 7B AWQ	0	5	4 GB

Best Alternatives to Typhoon 7B

Best Alternatives	Context / RAM	Downloads	Likes
...Nemo Instruct 2407 Abliterated	1000K / 24.5 GB	220	20
MegaBeam Mistral 7B 512K	512K / 14.4 GB	8470	54
SpydazWeb AI HumanAI RP	512K / 14.4 GB	16	1
SpydazWeb AI HumanAI 002	512K / 14.4 GB	18	1
...daz Web AI ChatML 512K Project	512K / 14.5 GB	12	0
MegaBeam Mistral 7B 300K	282K / 14.4 GB	3779	16
MegaBeam Mistral 7B 300K	282K / 14.4 GB	8018	17
Hebrew Mistral 7B 200K	256K / 30 GB	23	15
Astral 256K 7B	250K / 14.4 GB	9	0
Astral 256K 7B V2	250K / 14.4 GB	8	0

Note: green Score (e.g. "73.2") means that the model is better than scb10x/typhoon-7b.