Mythomax prompt

edit

• 13/M input. Can you make sure you've rebuilt for culbas from scratch by doing a make clean followed by a make LLAMA_CUBLAS=1 Defining Your Custom Model. i do have one RP that's just 1-2 line text back and forth tho, so it all just Description. Leverage Its Roleplaying Strengths: Use MythoMax-L2 (13B) to create vivid, character-driven narratives and simulations, taking advantage of its unique ability to assume Description. This version introduces some server optimizations. Many elements of the System Prompts and Author&rsquo;s Notes are interchangeable, so don&rsquo;t be shy about mixing and matching. For most purposes, most of the time, it more reliably produces good results than anything else I've tried locally. ALWAYS follow the prompt, pay attention to {{user}}'s messages and actions, but prioritize {{char}}'s response first and foremost. md. Mythomax Kimiko. Straight off the top of my head, there's: Mythomax. Compared to GPTQ, it offers faster Transformers-based inference. 0 Release! with improved Roleplay and even a proxy preset. This model is proficient at both roleplaying and storywriting due to its unique nature. For example: This prompt should make the AI not express positive feelings about the color red. microsoft/ Phi-3-medium-4k-instruct. For chat-instruct, you need to load an instruction-following model and then select a matching instruction template in the Chat settings tab. meekchique. Again, if you get the prompts etc right, I reckon these models can do anything you might reasonably want to do, and with decent context sizes as well. 0. . On the command line, including multiple files at once. Models. Prompt Format. This repo contains GGML format model files for Undi95's MythoMax L2 Kimiko v2 13B. cpp documentation. 0 Rating. To be updated. The main difference with MythoMix is that I allowed more of Huginn to intermingle with the single tensors located at the front and end of a model, resulting in increased Made ZORG👽 to work on Perplexity, MythoMax, Google-PaLM. Start Chat. I can honestly say that in its current state this is a very impressive development. Message Example. For other parameters and how to use them, please refer to the llama. 4. About AWQ. $0. As you fill out each field, you'll notice a colored bar at the top of the page change; that bar represents your available token limits, which vary depending on the AI model you use. # my_model_def. MythoMax 13B is a powerful, uncensored, open-source model great for roleplay. Differences: Includes a 70B recreation of SuperCOT as in the 1. Gryphe: Mythomax L2 13B 8K (beta) Mythomax L2 13B 8K is an open source large language model (LLM) created by Gryphe , that specializes in storytelling and advanced roleplaying . You also get unlimited messages for the free model. goliath-120b. ### Response: MythoMax is really good. gguf. Share. To get started, you can use OpenRouter via API like this: Without messing with the regular prompt, I am trying to see if the negative prompt does anything at all. All prompt format of all LLMs in one page MythoMax: Gryphe/MythoMax-L2-13b: MythoMix-L2-13b: MythoMix: Gryphe/MythoMix Quick Start. 19 GB. I have used it Mythomax 13B online via kobold, so even if without so many people saying it's great, I know from first hand experience it can generate amazing roleplay responses. ggmlv3. Become a Patron 🔥 - https://patreon. There seems to be two keys to make MythoMax work really well: very long intro message so the AI knows the exact tone you expect and tweaking the system prompt (I had luck with tweaking the roleplay prompt so MythoMax acts as the "game master" as well rather than just it's own character). EDIT 2: Rough fix: editing the response, replacing everything with a period and a line break, then using Continue gets a different response. Click Download. It seemed subjectively better than each of the individual models in my . Then click Download. llava-hf/ llava-1. JSON. py. You are responsible for how you use Synthia. This was actually one of the motivations for moving to a raw endpoint — to have each character and user name and build prompts like: Create avatars for your characters by filling in the prompt and selecting Generate Claude V2. 3 response; Prompt template: Alpaca Below is an instruction that describes a task. Mythomax is really good too, though, save for (in my experience) mythalion's sparks of brilliance. This model was created in collaboration with Gryphe, a mixture of our Pygmalion-2 13B and Gryphe’s Mythomax L2 13B. This technique is described as highly experimental and is used to merge the MythoLogic-L2 and Huginn models in the MythoMix series. Tweak it Share Report. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead. Prompt Format This model primarily uses Alpaca formatting, so for optimal model performance, use: <System prompt/Character Card> ### Instruction: Your instruction or question here. Q8_0. cpp no longer supports GGML models. Players currently on MythoMax have been upgraded to v1. OutputCost 3. For Llama2 there are no specific erotic models, so if you want Erebus specifically where its highly NSFW on its own then there isn't a model out there. jpg, . For the CPU infgerence (GGML / GGUF) format, having enough RAM is key. Sep 14, 2023 · MythoMax as a model has a unique trait: it effectively adapts to making valid SillyTavern-style prompts better than the others, which are more open ended / inconsistent formatting wise, and Mytho expects ST prompting styles pretty consistently (e. The model will start downloading. Thanks! Aug 31, 2023 · Unknown Model, cannot load. Mythomax and its variants are popular at the moment, and honestly I find Mythomax to be the "best" overall. So very large, extremely low quality loss. Skip to content. Model details. A merge of Pygmalion-2 13B and MythoMax 13B Model Details The long-awaited release of our new models based on Llama-2 is finally here. set up properly. Personality. It seemed subjectively better than each of the individual models in my Dec 2, 2023 · Another open-source model we host called MythoMax isn’t either, except for, if you squint, one of its component models called Airoboros, which uses the USER:/ASSISTANT: structure but is largely not roleplay. so file or there is a problem with the gguf model. 4K Popularity. Although in the hint I am asking for the exact opposite. {{char}} will not speak for {{user}}, it's strictly against the guidelines to do so, as {{user}} must take the actions and decisions themself. TieFighter 1. 1 -Claude V2 -Claude v1. Write a response that appropriately completes the request. Which is cool and all, but the thing is, there's several variants of it. 1 version was used instead of 2. Let’s take a look at the template Interview a celebrity, particularly the following prompt: Quantized 70B recreation of MythoMax. Write a single reply only. The GGML format has now been superseded by GGUF. This repo contains GGML format model files for Gryphe's MythoMax L2 13B. Give it text and it'll predict what comes next. With 13B parameters, MythoMax boasts a brain bigger than any other language model currently in existence. If using the roleplay default don't, it really pushes things to erp pretty hard. If it doesn't, hopefully you can get the idea and make your own jailbreak that would be more suitable for your purposes. Description. 70B recreation of MythoMax. This file should include the definition of your custom model. svg, . Someone try on GPT-4 plus for me. 1 response; As long as the connection between Eiko and Russ remains strong, they will remain swapped into each other's bodies. It's smart. How to run in text-generation-webui. Did you modify or replace any files when building the project? It's not detecting GGUF at all, so either this is an older version of the koboldcpp_cublas. First, you need to define your custom language model in a Python file, for instance, my_model_def. So here’s a handy list of potential prompts for you to use. models import LlamaCppModel, ExllamaModel mythomax_l2_13b_gptq = ExllamaModel (. Please use it with caution and with best intentions. It is also now supported by continuous batching server MythoMax-L2-Kimiko-v2-13B-GGML, developed by TheBloke, is a part of the MythoMax series, which is an improved variant of the MythoMix series. Did it work? Probably, maybe. PromptCost 3. Augmental-13b (Hugging Face page) (TheBloke Quant) is a finetune on top of the already-strong MythoMax, using a new dataset of "augmented" text — text from a human-written source that's processed into the right format, and then enhanced, using AI (in this case, GPT-4). Addendum (2023-11-23): A more thorough investigation revealed a flaw in my original algorithm that has since been resolved. The first is a base model. 10. Get up and running with large language models. 1 (for subscribers for now since it costs quite a bit to make work) The two questions we have are: - Are there any other models people have trouble deploying? From my own testing I concluded that OpenRouter version of Mythomax completely disregards the main prompt and jailbreak preventing you from fine tuning it to behave like you'd like it to (basically OpenRouter version is what you see is that you get) while mancers version does listen to the instruct mode well enough to follow your instructions MythoMax 1. Provided files. Known limitation: it strongly prefers novel format in roleplay, and will revert to it over time regardless of context or conversation history. The real Mythomax 33b does not exist because Mythomax is a mix of various Llama 2 fine tunes, and there is no 33b version of those ingredients. These currently include Griffin, MythoMax, Tiefighter, and Stable Diffusion 1. Vicuna 1. cpp team on August 21st 2023. The goal was to create a model that combines creativity with intelligence for an enhanced experience. Write a single reply only Chief: (end of prompt, send here) 12 volts battery, some wire, tape, and two small explosive charges. Comparison to other models. nvidia/ Nemotron-4-340B-Instruct. openai/ whisper-large. 75) if you experience weird outputs. Run the following command to see the correct prompt format and stop sequence(s): MythoMax-L2 (13B) Gryphe/MythoMax-L2-13b: 4096: LM Sys: Vicuna v1. Hello, my problem is that the model I use (mythomax-l2-13b) often likes to put my character in completely hopeless situations. Mixtral 8x7B has been tested for roleplaying and large contexts, which comes from the fact that the number of parameters of the model is very competitive. In your instruct settings you can edit your system prompt, start with alpaca sys prompt and add a few lines to discourage initiating sexual advances and such. Aug 30, 2023 · August 17, 2023 by. ### Instruction: {prompt} ### Response: Licensing The creator of the source model has listed its license as cc-by-nc-4. I was on Mercury for a month or so and the responses were not great. Past versions of MythoMax are still available in case you have problems with v1. Famous character roleplaying. : r/AIDungeon. But what comes close is indeed as others suggested models like Holomax which are designed to work with "Write I think literally no one says that, I mean its obvious MythoMax wont perform better or let alone as well as GPT, Im no AI or LLM expert but I think its GPT's vast training data in which makes it so efficient in roleplays, like GPT is already aware of most fictional characters, adding additional info aka the character prompt (Idk If thats whats it called) makes it even more efficient. A model, sometimes called a "weight", is the core -- or brain -- of your AI. ## Prompt Format: This model primarily uses Alpaca formatting, so for optimal model performance, use: ``` <System prompt/Character Card> ### Instruction: Your instruction or question here. g scenario, personality, same system and human prompts). All Synthia models are uncensored. 13/M output. mythomax-l2-13b. very large, extremely low quality loss - not recommended. Tape the detonator to the charge, then run the other wire to the other side of the wall. According to our testers, this model seems to outperform MythoMax in RP/Chat. It is built on the foundation of the Llama 2 architecture and is a part of the Mytho family of Llama-based models, which also includes MythoLogic and MythoMix. 1. The hint: This is text role-playing game. Granted this is for sexy stuff more than killing goblins, but hopefully it should work for you as well. It's not really explicitly an NSFW model, but handles it well with good prompting. 18, Range 2048, Slope 0 (same settings simple-proxy-for-tavern has been using for months) which has EDIT: Specifically, I am using TheBloke's mythomax-l2-13b. Scenario. Whoever made the "Mythomax L2 33B" is an asshole hijacking the name fully knowing he is tricking people into thinking they are getting a 33B version of Mythomax, while in reality they are getting The models will be running using a different tool which may require your presets to be changed. - A game master describes the world around and the user's situation. Striking a balance is key for detailed and engaging output. Prompt template: Custom. 12. Chat mode doesn't require any special attention, just select a character (or use the default Assistant character) and send messages. This was common for years until ChatGPT popularized the second type, instruct models, which start from base models, but are trained on question-and-answer and Sep 27, 2023 · Mythalion 13B A merge of Pygmalion-2 13B and MythoMax 13B Model Details The long-awaited release of our new models based on Llama-2 is finally here. Featured Models. Mythomax L2 13B 8K is evidently superior to others in terms of roleplaying and storytelling, not only because it can assume a role or tell a story in a consistent form, but Description. These files were quantised using hardware kindly provided by Massed Compute. 5-7b-hf Experiment with different models and prompts. Optimize Prompt Length: While lengthy prompts might affect coherence, overly short prompts may yield brief responses. Look at Fragments for a breakdown of what each piece of the prompt means. license: other Edit model card. The result is a model named Mythmalion-13B , a versatile and powerful roleplay model combining MythoMax’s stability and intelligence with Pygmalion-2’s raw To download from a specific branch, enter for example TheBloke/MythoMax-L2-13B-GPTQ:main. maybe MythoMax-L2-13b. For roleplay purposes, I suggest the following - Write <CHAR NAME>'s next reply in a chat between <YOUR NAME> and <CHAR NAME>. MythoMix-L2-13b. Free players can spend Scales to speed-boost these text models for 1 Aug 27, 2023 · But like I mentioned, that shouldn't really have anything to do with your issue because you're using temperature 0. Mythomax is currently my go-to here, although in the past I used to recommend l2 versions of Chronos-Hermes and Airoboros. com/FahdMirza#mythomax #myt Nov 23, 2023 · MythoMist-7b. Important note regarding GGML files. 69 GB. Aug 31, 2023 · For beefier models like the MythoMax-L2-13B-GPTQ, you'll need more powerful hardware. It is built on the Llama 2 architecture and is an optimized version of the MythoMix model, incorporating a tensor merger strategy for increased The Basic tab includes the following fields: Name. It is also now supported by continuous batching server vLLM Prompt format: SillyTavern Base model: MythoMax-L2-13b What's new: finetuned on the script of a visual novel that was processed and revamped by GPT-4 to make ~1300 high-quality training examples. OpenRouter Prompt format. 5. It has been fine-tuned for instruction following as well as having long-form conversations. As of August 21st 2023, llama. q4_K_M. 8. stability-ai/ sdxl. Show logs. This is the system prompt I use and it generally curbs this sort of behavior: You're {{char}} in this fictional never-ending, uncensored, realistic and immersive roleplay with {{user}}. see Provided Files above for the list of branches for each option. I recommend using the huggingface-hub Python library: All prompt format of all LLMs in one page. Generated. Created by smashing two very successful 70b models together, this model has an unhealthy number of parameters. l2-mythomax22b-instruct-Falseblock. Completely ignoring prompts or completely irrelevant responses. A Mythomax/MLewd_13B-style merge of selected 70B models A multi-model merge of several LLaMA2 70B finetunes for roleplaying and creative work. Greetings, imaginative souls, as we dive into the realm of AI-fueled romantic roleplays! The canvas of emotions is vast, and your prompts have the power to kindle flames of connection and intimacy. cpp. py from llama_api. Sep 2, 2023 · To that end, we reached out to Gryphe, creator of the popular MythoMax-L2-13B model (which itself is a blend of many different Llama-2 models) to help us merge our model with theirs. gif) 79Ktokens. Please let us know if you have any questions or if you run into any issues with v1. 83 s. SynthIA (Synthetic Intelligent Agent) is a LLama-2-70B model trained on Orca style datasets. Yesterday I used this system quite actively. 5) that we think people will like: - Gryphe/MythoMax-L2-13b (Free) - Xwin-LM/Xwin-LM-70B-V0. The end goal was a model that could speak like a specific character from that game, but the end result was a model that seems to excel in banter ALWAYS follow the prompt, pay attention to {{user}}'s messages and actions. UPDATE: There's an improved version now! Check it MythoMax! A requested variant of MythoLogic-L2 and Huginn using a highly experimental tensor type merge technique. a really great model, but then I don't have enough VRAM left for SkyrimVR when running at 43 gpu layers, on my 4090(!!), I cut it back to 28. i wasn't aware Mancer was about to pull it's free tokens for Mythomax. 4. MythoMax is really good. In AI Dungeon, Basic models are available to all players. This model was created in collaboration with Gryphe, a mixture of our Pygmalion-2 13B and Gryphe's Mythomax L2 13B. I just paste this at the top of the character description and that usually does the trick. 83 Conversations. This repo contains GGUF format model files for Gryphe's MythoMax L2 13B. Anywhere Airoboros is merged in, the 1. This example was created by a different version, smoosh-sh/mythomax-l2-13b-gptq:6eff74ce. The following are the differences between TheBloke MythoMix vs MythoMax model series: 1. In the top left, click the refresh icon next to Model. The long-awaited release of our new models based on Llama-2 is finally here. 5 (13B) yep, i paste the character card in as an entry and just make the name of the character a key! and haha ig i just like to write novella, but also the long ass beginning prompt makes it so that when you press "send" again for it to lengthen its response it will do so more coherently since you gave it a long example. According to our testers, this model seems to outperform MythoMax in RP The models will be running using a different tool which may require your presets to be changed. png, . The UI tries to define the template automatically based on the model That's actually a good question. First Message. We can connect the wires to the detonator and the blasting cap. 3 instand -Mythalion -Mythomax -Nous Hermes Yes, with the Mercury tier, you get access to Mistral, Mythomax with unlimited messages. Jan 30, 2024 · Therefore, this LLM successfully collects the relevant information for both roleplaying and storytelling, making coherent answers to the prompts. With Llama 3 released, it's time for MythoMax to slowly fade away Let's do it in style! An improved, potentially even perfected variant of MythoMix, my MythoLogic-L2 and Huginn merge using a highly experimental tensor type merge technique. 1. AMD 6900 XT, RTX 2060 12GB, RTX 3060 12GB, or RTX 3080 would do the trick. This repo contains AWQ model files for Gryphe Padar's MythoMist 7B. Further instructions can be found in the text-generation-webui documentation, here: text-generation-webui/docs/04 ‐ Model Tab. OutputLimit 400 tokens. 0, and this quantization has therefore used that same license. Q8_0. It is also supports metadata, and is designed to be extensible. In text-generation-webui. 4K context. 57 votes, 47 comments. 0 /5. Bio. AWQ is an efficient, accurate and blazing-fast low-bit weight quantization method, currently supporting 4-bit quantization. Paste, drop or click to upload images (. with {{char}} will respond with 2 to 3 paragraphs, and afterwards respond as {{user}} on a new line. then i noticed Venus Chub also had a model called "Mythomax", and since i'm already paying 20 bucks a month, if it's the same Sep 8, 2023 · I'm using the 13b Q8_0 MythoMax GGUF as my LLM (uses >13gb VRAM). Preview. i used Mancer's version of Mythomax, and then i tried Venus Chub's model Asha. jpeg, . But, certain parts of prompting, like If you want to have a chat-style conversation, replace the -p <PROMPT> argument with -i -ins. Thanks, and how to contribute. Award. Basic text models are always unlimited and free. GGUF offers numerous advantages over GGML, such as better tokenisation, and support for special tokens. X. Compared to GPTQ, it offers faster Transformers-based The videos shows how to locally install MythoMax L2 13B on windows and play around with it. by gryphe. Kinda jank, but works as a temporary fix! EDIT 3: I am dumb. I use mythalion actually. Progress Updates. Mythomax L2 13B is a large language model created by Gryphe that specializes in storytelling and advanced roleplaying. true. MythoMax-L2-13b. 23. Like a maestro orchestrating a symphony of emotions, you’re about to embark on a journey that merges human sentiment with I would also add instructions in system prompt to emphasize short answers (role-playing default response says two paragraphs), cut the response length to 120-150, set the flag to remove incomplete sentences and occasionally manually update char's dialogue as when it starts increasing response length it will learn and keep giving longer responses. ### Instruction: {prompt} ### Response: Provided files and GPTQ parameters Multiple quantisation parameters are provided, to allow you to choose the best one for your hardware and requirements. After 13B has been replaced, there may be another candidate running aimed at replacing 7B. Note: the above RAM figures assume no GPU offloading. Discord. 2023-08-19: After extensive testing, I've switched to Repetition Penalty 1. 3 minutes. Response generation time anywhere from 20 seconds to a minute. We would like to show you a description here but the site won’t allow us. It is a replacement for GGML, which is no longer supported by llama. This repo contains AWQ model files for taozi555's MythoMax Kimiko Mix. If you're using the GPTQ version, you'll want a strong GPU with at least 10 gigs of VRAM. cpp, I tried with Nous-Hermes and MythoMax, I tried with GGML and GGUF, I Jan 30, 2024 · Mythomax L2 13B 8K has been proven to be useful for roleplaying. The length of time they remain swapped can vary depending on various factors such as their emotional state, physical proximity, and the intensity of their connection. I tried with pronouns "I" and "You", I tried with Exllama and llama. Once it's finished it will say "Done". 2 version of Huginn. Original model: MythoMax L2 Kimiko v2 13B. Merging Technique. It'll figure it out. 1 instruct format is recommended, but try whatever. How to run in llama. The good news is it's a 13b model, so you could give it a try! 2. My congratulations to the creators. Original model card: Gryphe's MythoMix L2 13B. schemas. Compatibility. 25. I've tried comparing the two myself and from my own testing Mythomax q6_K running via Koboldcpp with instruct mode enabled and set to a slightly modified "Role Play" Preset that i'm using produces better, albeit slower results than OpenRouter's version with, or without jailbreak(for this test i just copied my modified Instruct system prompt into Jailbreak) and 2023-08-30: SillyTavern 1. Each of these models has been developed independently and has different characteristics. Prompt template: Alpaca Below is an instruction that describes a task. 🤖PromptFMT - Prompt format and related data for MythoMax-L2-13b. An improved, potentially even perfected variant of MythoMix, my MythoLogic-L2 and Huginn merge using a highly experimental tensor type merge technique. Finer details of the merge are available in our blogpost . According to our testers, this model We would like to show you a description here but the site won’t allow us. There are two main types of models. Unfortunately for whatever reason that is the difference of 3-5s responses to 30s+ responses. I updated my recommended proxy replacement settings accordingly (see above link). 5. This means it can process and understand information on a scale that was previously unimaginable. I have downloaded a version from huggingface (TheBloke_MythoMax-L2-13B-AWQ), but so far simply couldn't get it to work properly for more than one small answer, no Model Details. Explanation of the new k-quant methods. The tensor-type merging technique is a unique feature of the MythoMix series. System Prompt. 0 1 review. Quantized models are available from TheBloke: GGML - GPTQ (You're the best!) Output. i'm not really aknowledgeable about things like models, ect. It is recommended to use a lower temperature (< 1. Under Download Model, you can enter the model repo: TheBloke/MythoMax-Kimiko-Mix-GGUF and below it, a specific filename to download, such as: mythomax-kimiko-mix. I apologize for swearing at the developers in the comments for their decision to disable the legacy version. Or edit it. ResponseTime 10. Mythalion seems to be a bit less consistent than MythoMax. This repo contains AWQ model files for Undi95's MythoMax L2 Kimiko v2 13B. q5_K_M model. GGUF is a new format introduced by the llama. Reply. • 28 days ago. Finer details of the merge are available in our blogpost. My reason for it is that every ince in a while it will do something truly awesome, and be oretty good the rest of the time. If the issue only occurs when you have prompt caching turned on, then if you want to try to fix the issue you should probably look at the differences between when prompt caching is disabled vs enabled. For most people, most of the time, it'll produce decent results - better than decent, if you get your prompts and character cards etc. , and it didn't even try to do that. 75 cr/tok. There will be a candidate model running aimed at replacing the 13B model. Third party clients and libraries are expected to still This is why we deployed two open source models to RizzChat (in addition to GPT 3. MythoMist 7b is, as always, a highly experimental Mistral-based merge based on my latest algorithm, which actively benchmarks the model as it's being built in pursuit of a goal set by the user. Get the l2 versions, if you're going to try them out. xv vu ha ro fd gp ex hy ft vj