Qwen2.5 Coder Instruct 7b (Q6_K)

Qwen2.5 Coder Autocomplete 1.5b (Q6_K)

Qwen2.5 Coder Instruct 3b (Q6_K)

Qwen2.5 Coder Instruct 14b (Q6_K)

Gemma3 4b (Q6_K)

Gemma3 12b (Q6_K)

Gemma3 Enhanced 12b

Nvidia AceInstruct 7B (Q6_K)

Nomic Embed Text v2

Nomic Embed Code

OpenThinker2 7B (Q6_K)

Fast Apply v0.13.2

Qwen2.5-VL 3B Instruct

Devstral

BGE Reranker v2 M3

Reranker is disigned in cross-encoder architecture that takes the query and text at the same time and directly output their score of similarity.

Nvidia AceReason Nemotron 14B (Q6_K)

Fast Apply 1.5B v1.0

QwQ 32B (Q4_K_M)

DeepSeek-R1 0528 Qwen3 8B (Q6_K)

Phi-4 (Q6_K)

Qwen2.5-VL 7B Instruct

Codestral

Cutting-edge language model for coding. With the second version released January 2025, Codestral specializes in low-latency, high-frequency tasks such as fill-in-the-middle (FIM), code correction and test generation.