Abacusai/llama-3-smaug-8b Autotrain compatible Base model:abacusai/llama-3-sm... Base model:cognitivecomputatio... Base model:dreamgen-preview/op... Base model:merge:abacusai/llam... Base model:merge:cognitivecomp... Base model:merge:dreamgen-prev... Base model:merge:weyaxi/einste... Base model:weyaxi/einstein-v6.... Cognitivecomputations/dolphin-... Dreamgen-preview/opus-v1.2-lla... Endpoints compatible Frankenmoe Merge Mergekit Mixtral Moe Region:us Safetensors Sharded Tensorflow Weyaxi/einstein-v6.1-llama3-8b
Skyro 4X8B Benchmarks
ARC: 61.26 vs 96.7 (so35)-36.6%
HellaSwag: 82.38 vs 95.3 (gpt4)-13.6%
MMLU: 66.67 vs 88.3 (so35)-24.5%
TruthfulQA: 50.15 vs 59 (gpt4)-15%
WinoGrande: 77.66 vs 87.5 (gpt4)-11.2%
GSM8K: 60.2 vs 96.4 (so35)-37.6%
LLME Score: 0.22868
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
Skyro 4X8B Parameters and Internals
| LLM Name | Skyro 4X8B |
| Repository 🤗 | https://huggingface.co/saucam/Skyro-4X8B |
| Base Model(s) | |
| Model Size | 24.9b |
| Required VRAM | 50 GB |
| Updated | 2024-09-07 |
| Maintainer | saucam |
| Model Type | mixtral |
| Model Files | |
| Model Architecture | MixtralForCausalLM |
| License | apache-2.0 |
| Context Length | 8192 |
| Model Max Length | 8192 |
| Transformers Version | 4.40.1 |
| Tokenizer Class | PreTrainedTokenizerFast |
| Padding Token | <|begin_of_text|> |
| Vocabulary Size | 128256 |
| Torch Data Type | bfloat16 |
Best Alternatives to Skyro 4X8B
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| L3.1 ClaudeMaid 4x8B | 128K / 50.1 GB | 51 | 7 |
| Llama Salad 4x8B V3 | 8K / 50.1 GB | 11 | 4 |
| L3 SnowStorm V1.15 4x8B B | 8K / 49.9 GB | 21 | 9 |
| ...ama 3 Aplite Instruct 4x8B MoE | 8K / 50 GB | 58 | 38 |
| L3 Arcania 4x8b | 8K / 50 GB | 32 | 22 |
| L3 SnowStorm 4x8B | 8K / 49.9 GB | 4 | 11 |
| L3 SnowStorm V1.15 4x8B A | 8K / 49.9 GB | 7 | 1 |
| ChaoticSoliloquy 4x8B | 8K / 50 GB | 15 | 25 |
| Llama4Some SOVL 4x8B L3 V1 | 8K / 50 GB | 5 | 2 |
| L3 ChaoticSoliloquy V1.5 4x8B | 8K / 49.9 GB | 7 | 8 |
Note: green Score (e.g. "73.2") means that the model is better than saucam/Skyro-4X8B .
Rank the Skyro 4X8B Capabilities
🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟
What open-source LLMs or SLMs are you in search of? 35693 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v2024072803