Quantized versions of Prometheus 2 - an alternative of GPT-4 evaluation when doing fine-grained evaluation of an underlying LLM.
Seva Leonov PRO
vsevolodl
AI & ML interests
ML Ops, scalable Inference, quantization, long context, fine-tuning
Recent Activity
liked a model 2 days ago
Qwen/Qwen3.5-4B upvoted a collection 2 months ago
NVIDIA Nemotron v3 liked a model 2 months ago
Qwen/Qwen3-VL-Reranker-2BOrganizations
models 18
vsevolodl/qwen3-embed-v4-bashar_docs-192-hn
Sentence Similarity • 0.6B • Updated • 6
vsevolodl/qwen3-embed-v4-bashar_docs-192
Sentence Similarity • 0.6B • Updated • 3
vsevolodl/qwen3-embed-v3-bashar_docs-192
Sentence Similarity • 0.6B • Updated
vsevolodl/qwen3-embed-v3-bashar_docs-192-hn
Sentence Similarity • 0.6B • Updated
vsevolodl/Llama-3-8B-Instruct-Gradient-4194k-GGUF
8B • Updated • 24 • 1
vsevolodl/Yi-1.5-34B-Chat-GGUF
34B • Updated • 25 • 1
vsevolodl/falcon-11B-GGUF
11B • Updated • 36 • 1
vsevolodl/Yi-1.5-9B-Chat-GGUF
9B • Updated • 54 • 1
vsevolodl/Llama-3-70B-Instruct-Gradient-1048k-GGUF
Text Generation • 71B • Updated • 27 • 3
vsevolodl/Llama-3-8B-Instruct-Gradient-1048k-GGUF
Text Generation • 8B • Updated • 47 • 1