Autotrain compatible Conversational Endpoints compatible Instruct Qwen2 moe Region:us Safetensors Sharded Tensorflow
Model Card on HF 🤗: https://huggingface.co/ehristoforu/tmoe-v2
Tmoe V2 Benchmarks
MMLU Pro: 1.12
GPQA: 1.79
MUSR: 10.05
BBH: 2.06
IFEval: 19.03 vs 88 (so35)-78.4%
MATH Lvl 5: 0.23
LLME Score: 0.21741
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)
Tmoe V2 Parameters and Internals
| LLM Name | Tmoe V2 |
| Repository 🤗 | https://huggingface.co/ehristoforu/tmoe-v2 |
| Model Size | 1.5b |
| Required VRAM | 22.1 GB |
| Updated | 2025-09-17 |
| Maintainer | ehristoforu |
| Model Type | qwen2_moe |
| Instruction-Based | Yes |
| Model Files | |
| Model Architecture | Qwen2MoeForCausalLM |
| License | apache-2.0 |
| Context Length | 32768 |
| Model Max Length | 32768 |
| Transformers Version | 4.48.1 |
| Tokenizer Class | Qwen2Tokenizer |
| Padding Token | <|im_end|> |
| Vocabulary Size | 151936 |
| Torch Data Type | bfloat16 |
| Errors | replace |
Rank the Tmoe V2 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? 51544 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124