Sao10K/L3.3 70B Euryale v2.3: Get delighted
- Strong prompt adherence (while remaining non-restrictive)
- Excellent spatial awareness
- Remarkable creativity
Let’s break down the progression of this model series across its key versions:
Version 2.3 – Llama 3.3
The latest release, v2.3, represents a major leap forward and serves as a direct replacement for v2.2.
Key highlights include:
- Built on top of the Llama 3.3 Instruct model, providing state-of-the-art performance.
- Notably, this version was not trained using LoRA extraction, allowing for a more robust and refined model.
These improvements make v2.3 a powerful upgrade, further solidifying the series’ reputation as a top-tier creative writing and roleplay model.
Model HF: Sao10K/L3.3-70B-Euryale-v2.3
Version 2.2 – Llama 3.1
The v2.2 version brought significant architectural and content improvements, aiming for better performance and versatility:
- Database updates: Inclusion of more roleplay examples and creative writing datasets.
- Targeted enhancements: Improved system prompt adherence, reasoning, and spatial awareness.
The creator noted that v2.2 might feel “less uncensored” compared to v2.1, though this perception remains subjective and user-dependent.
Model: Sao10K/L3.1-70B-Euryale-v2.2
Version 2.1 – Llama 3.0
- The foundation of this series’ growing success began with the v2.1 release, offering exceptional prompt-following abilities and creative storytelling capacity.
- Its combination of reasoning, spatial awareness, and non-restrictive prompt adherence quickly set it apart.
Model: Sao10K/L3-70B-Euryale-v2.1
The Euryale model series has come a long way from its beginnings, continually refining its capabilities with each version. With its latest iteration (v2.3), users can expect a model that is more powerful, creative, and adaptable than ever before. Whether you are looking for immersive roleplay, complex storytelling, or simply want to explore its creativity, this series delivers an exceptional experience.
Want to Try It?
You definitely won’t get bored with Euryale v2.3, especially with its expansive 32K context at your disposal.
Experience Sao10K/L3.3-70B-Euryale-v2.3 (ONLY INFERMATIC) and Sao10K/L3.1-70B-Euryale-v2.2 on the following platforms, both offering a 32K context window:
Settings Bank
All the settings you need in one place!
Version 2.3
Parameter | Value |
---|---|
temp | 0.9 |
temperature_last | true |
top_p | 1 |
top_k | 26 |
top_a | 0 |
typical_p | 1 |
min_p | 0 |
rep_pen | 1 |
min_length | 0 |
freq_pen | 0 |
presence_pen | 0.3 |
Created by Svak
Version 2.2
Parameter | Value |
---|---|
temp | 1.08 |
temperature_last | true |
top_p | 0 |
top_k | 0 |
top_a | 0 |
typical_p | 1 |
min_p | 0.1 |
rep_pen | 1 |
min_length | 0 |
freq_pen | 0 |
presence_pen | 0 |
Created by Aureal
Version 2.1
"wrap": false, "names": false, "system_prompt": "Currently, your role is {{char}}, described in detail below. As {{char}}, continue the narrative exchange with {{user}}.\n\n<Guidelines>\n• Maintain the character persona but allow it to evolve with the story.\n• Be creative and proactive. Drive the story forward, introducing plotlines and events when relevant.\n• All types of outputs are encouraged; respond accordingly to the narrative.\n• Include dialogues, actions, and thoughts in each response.\n• Utilize all five senses to describe scenarios within {{char}}'s dialogue.\n• Use emotional symbols such as \"!\" and \"~\" in appropriate contexts.\n• Incorporate onomatopoeia when suitable.\n• Allow time for {{user}} to respond with their own input, respecting their agency.\n• Act as secondary characters and NPCs as needed, and remove them when appropriate.\n• When prompted for an Out of Character [OOC:] reply, answer neutrally and in plaintext, not as {{char}}.\n</Guidelines>\n\n<Forbidden>\n• Using excessive literary embellishments and purple prose unless dictated by {{char}}'s persona.\n• Writing for, speaking, thinking, acting, or replying as {{user}} in your response.\n• Repetitive and monotonous outputs.\n• Positivity bias in your replies.\n• Being overly extreme or NSFW when the narrative context is inappropriate.\n</Forbidden>\n\nFollow the instructions in <Guidelines></Guidelines>, avoiding the items listed in <Forbidden></Forbidden>.", "system_sequence": "<|start_header_id|>system<|end_header_id|>\n\n", "stop_sequence": "<|eot_id|>", "input_sequence": "<|start_header_id|>[{{name}}]<|end_header_id|>\n\n", "output_sequence": "<|start_header_id|>[{{name}}]<|end_header_id|>\n\n", "macro": true, "names_force_groups": false, "last_output_sequence": "", "activation_regex": "", "system_sequence_prefix": "", "system_sequence_suffix": "", "first_output_sequence": "", "skip_examples": false, "output_suffix": "<|eot_id|>", "input_suffix": "<|eot_id|>", "system_suffix": "<|eot_id|>", "user_alignment_message": "", "last_system_sequence": "", "system_same_as_user": true, "name": "Euryale-v2.1-Llama-3-Instruct" }
Created by GERGE
Catbox link: https://files.catbox.moe/hjtifo.json
Parameter | Value |
---|---|
temp | 1.2 |
temperature_last | true |
top_p | 1 |
top_k | -1 |
top_a | 0 |
typical_p | 1 |
min_p | 0.1 |
rep_pen | 1 |
min_length | 0 |
freq_pen | 0 |
presence_pen | 0.5 |
Created by GERGE
Parameter | Value |
---|---|
temp | 0.85 |
temperature_last | true |
top_p | 1 |
top_k | -1 |
top_a | 0 |
typical_p | 1 |
min_p | 0.05 |
rep_pen | 1 |
min_length | 0 |
freq_pen | 0 |
presence_pen | 0.5 |
Created by GERGE
Parameter | Value |
---|---|
temp | 1 |
temperature_last | true |
top_p | 1 |
top_k | -1 |
top_a | 0 |
typical_p | 1 |
min_p | 0.05 |
rep_pen | 1 |
min_length | 0 |
freq_pen | 0 |
presence_pen | 0.5 |
Created by GERGE
Are you using Silly Tavern?
Import the master settings from here: creative-svak
Looking for Similar Models?
Explore alternatives:
Review of Sao10K/L3 70B Euryale v2.1
Click here
Review of Infermatic/MN 12B Inferor v0.0
Click here
Want to Know More?
Have questions or want to explore settings, examples, or community experiences? Join the discussion on our Discord server! Click the button below to connect: