exaone-deep

EXAONE Deep exhibits superior capabilities in various reasoning tasks including math and coding benchmarks, ranging from 2.4B to 32B parameters developed and released by LG AI Research.

2.4b 7.8b 32b

Hızlı Kurulum (Ollama kuruluysa)

ollama run exaone-deep

Ollama kurulu değil mi? ollama.com/download — Windows, macOS ve Linux için ücretsiz. İlk çalıştırmada model indirilir, sonrası tamamen çevrimdışıdır.

Varyantlar

Boyut büyüdükçe kalite artar, donanım ihtiyacı yükselir. Başlangıç için küçük varyantı deneyin.

Etiket	Boyut	Bağlam	Girdi	Komut
latest	4.8GB	32K	Text	`ollama run exaone-deep:latest`
2.4b	1.6GB	32K	Text	`ollama run exaone-deep:2.4b`
7.8b	4.8GB	32K	Text	`ollama run exaone-deep:7.8b`
32b	19GB	32K	Text	`ollama run exaone-deep:32b`
2.4b-q4_K_M	1.6GB	32K	Text	`ollama run exaone-deep:2.4b-q4_K_M`
2.4b-q8_0	2.8GB	32K	Text	`ollama run exaone-deep:2.4b-q8_0`
2.4b-fp16	5.3GB	32K	Text	`ollama run exaone-deep:2.4b-fp16`
7.8b-q4_K_M	4.8GB	32K	Text	`ollama run exaone-deep:7.8b-q4_K_M`
7.8b-q8_0	8.3GB	32K	Text	`ollama run exaone-deep:7.8b-q8_0`
7.8b-fp16	16GB	32K	Text	`ollama run exaone-deep:7.8b-fp16`
32b-q4_K_M	19GB	32K	Text	`ollama run exaone-deep:32b-q4_K_M`
32b-q8_0	34GB	32K	Text	`ollama run exaone-deep:32b-q8_0`
32b-fp16	64GB	32K	Text	`ollama run exaone-deep:32b-fp16`

Model Detayları ve Benchmarklar (kaynak: ollama.com)

EXAONE Deep exhibits superior capabilities in various reasoning tasks including math and coding benchmarks, ranging from 2.4B to 32B parameters developed and released by LG AI Research.

Evaluation results show that:

EXAONE Deep 2.4B outperforms other models of comparable size
EXAONE Deep 7.8B outperforms not only open-weight models of comparable scale but also a proprietary reasoning model OpenAI o1-mini
EXAONE Deep 32B demonstrates competitive performance against leading open-weight models.

References

Hugging Face