Get a weekly rundown of the latest AI models and research... subscribe!

Goliath 120b



The Goliath-120b is an auto-regressive, causal Language Model (LM) created by merging two fine-tuned Llama-2 70B models. It works as a text-to-text AI, capable of generating comprehensive written outputs based on given prompts. The model takes various parameters such as top_k, top_p, prompt, temperature, max_new_tokens, prompt_template, presence_penalty, and frequency_penalty. The output schema is mainly text content responding to the given prompt. For instance, when given a prompt about Parmigiano Reggiano, the model generates a blog post-like text coherent with the specified theme.

Use cases

The Goliath-120b is an AI model that transforms text inputs into elaborate text outputs. This model could be utilized in various real-world scenarios. One such use could be in content creation for blogs or online magazines. The model can generate entire blog posts when given a prompt, making it a helpful resource for writers seeking inspiration or requiring assistance with content generation. For example, it can be asked to write a blog post about a specific topic, such as 'parmigiano reggiano,' providing helpful tips on the subject. Furthermore, it can be utilized to develop a virtual personal assistant or chatbot capable of generating detailed responses. With its given 'prompt-template', the model can partake in an organic-like conversational style with users, making it invaluable for customer service oriented businesses. Similarly, this technology might be incorporated into learning management systems to provide elaborate explanations or responses to student queries, enhancing the overall learning experience. Finally, given its capacity to express dissatisfaction with inferior products, in this case, poor quality American parmesan, it might find applications in reviewing or comparing products, therefore contributing to consumer awareness in product selection. In general, its capacity to formulate substantial and detailed text outputs could prove useful across various sectors such as education, customer service, content creation, and reviews.


Cost per run
Avg run time

Creator Models

Causallm 14b$?935
Stablecode Completion Alpha 3b 4k$?154
Yi 6b$?48
Yi 34b$?102

Similar Models

No similar models found

Try it!

You can use this area to play around with demo applications that incorporate the Goliath 120b model. These demos are maintained and hosted externally by third-party creators. If you see an error, message me on Twitter.

Currently, there are no demos available for this model.


Summary of this model and related resources.

Model NameGoliath 120b

An auto-regressive causal LM created by combining 2x finetuned Llama-2 70B ...

Model LinkView on Replicate
API SpecView on Replicate
Github LinkNo Github link provided
Paper LinkNo paper link provided


How popular is this model, by number of runs? How popular is the creator, by the sum of all their runs?

Model Rank
Creator Rank


How much does it cost to run this model? How long, on average, does it take to complete a run?

Cost per Run$-
Prediction Hardware-
Average Completion Time-