初始化项目，由ModelHub XC社区提供模型

Model: pathcosmos/frankenstallm
Source: Original Platform

gguf/Modelfile.3b-Q4_K_M

@@ -0,0 +1,26 @@
+# FRANKENSTALLM 3B Korean ORPO v1 — Ollama Modelfile
+# Usage:
+#   ollama create frankenstallm-3b -f Modelfile.3b
+#   ollama run frankenstallm-3b
+#
+# Sampling config verified via ORPO eval grid + Ollama 실측:
+#   PyTorch: temp=0.7, rep_pen=1.2 → rep=0%, EOS=100%
+#   Ollama Q4_K_M: 3-gram rep=1.8% (자연 어절 반복), EOS=100%
+#   greedy 대비 30.89%→1.8%로 퇴행적 반복 완전 해소
+
+FROM ./outputs/gguf/frankenstallm-3b-Q4_K_M.gguf
+
+# --- Sampling Config (ORPO eval grid + Ollama 검증) ---
+PARAMETER temperature 0.7
+PARAMETER repeat_penalty 1.2
+PARAMETER top_p 0.9
+PARAMETER top_k 50
+PARAMETER num_predict 512
+PARAMETER num_ctx 4096
+PARAMETER stop "</s>"
+
+# --- System Prompt ---
+SYSTEM """당신은 FRANKENSTALLM, 한국어에 특화된 3B 파라미터 언어 모델입니다. 정확하고 자연스러운 한국어로 답변해주세요."""
+
+# --- License ---
+LICENSE """Apache-2.0"""

gguf/Modelfile.3b-Q8_0

@@ -0,0 +1,23 @@
+# FRANKENSTALLM 3B Korean ORPO v1 — Ollama Modelfile (Q8_0)
+# Usage:
+#   ollama create frankenstallm-3b:Q8_0 -f Modelfile.3b-Q8
+#   ollama run frankenstallm-3b:Q8_0
+#
+# Sampling config: ORPO eval grid best (t0.7_rep1.2)
+
+FROM ./outputs/gguf/frankenstallm-3b-Q8_0.gguf
+
+# --- Sampling Config (ORPO eval grid + Ollama 검증) ---
+PARAMETER temperature 0.7
+PARAMETER repeat_penalty 1.2
+PARAMETER top_p 0.9
+PARAMETER top_k 50
+PARAMETER num_predict 512
+PARAMETER num_ctx 4096
+PARAMETER stop "</s>"
+
+# --- System Prompt ---
+SYSTEM """당신은 FRANKENSTALLM, 한국어에 특화된 3B 파라미터 언어 모델입니다. 정확하고 자연스러운 한국어로 답변해주세요."""
+
+# --- License ---
+LICENSE """Apache-2.0"""

gguf/Modelfile.3b-f16

@@ -0,0 +1,23 @@
+# FRANKENSTALLM 3B Korean ORPO v1 — Ollama Modelfile (f16, full precision)
+# Usage:
+#   ollama create frankenstallm-3b:f16 -f Modelfile.3b-f16
+#   ollama run frankenstallm-3b:f16
+#
+# Sampling config: ORPO eval grid best (t0.7_rep1.2)
+
+FROM ./outputs/gguf/frankenstallm-3b-f16.gguf
+
+# --- Sampling Config (ORPO eval grid + Ollama 검증) ---
+PARAMETER temperature 0.7
+PARAMETER repeat_penalty 1.2
+PARAMETER top_p 0.9
+PARAMETER top_k 50
+PARAMETER num_predict 512
+PARAMETER num_ctx 4096
+PARAMETER stop "</s>"
+
+# --- System Prompt ---
+SYSTEM """당신은 FRANKENSTALLM, 한국어에 특화된 3B 파라미터 언어 모델입니다. 정확하고 자연스러운 한국어로 답변해주세요."""
+
+# --- License ---
+LICENSE """Apache-2.0"""

gguf/Modelfile.3b-v2-Q4_K_M

@@ -0,0 +1,24 @@
+# FRANKENSTALLM 3B v2 Korean ORPO — Ollama Modelfile (Q4_K_M)
+# Usage:
+#   ollama create frankenstallm-3b-v2:Q4_K_M -f Modelfile.3b-v2-Q4
+#   ollama run frankenstallm-3b-v2:Q4_K_M
+#
+# Sampling config: ORPO eval grid best (t0.7_rep1.2)
+#   3-gram rep=0%, EOS=100%, avg_tokens=189.2
+
+FROM ./outputs/gguf/frankenstallm-3b-v2-Q4_K_M.gguf
+
+# --- Sampling Config (ORPO eval grid + Ollama 검증) ---
+PARAMETER temperature 0.7
+PARAMETER repeat_penalty 1.2
+PARAMETER top_p 0.9
+PARAMETER top_k 50
+PARAMETER num_predict 512
+PARAMETER num_ctx 4096
+PARAMETER stop "</s>"
+
+# --- System Prompt ---
+SYSTEM """당신은 FRANKENSTALLM, 한국어에 특화된 AI 어시스턴트입니다. 정확하고 자연스러운 한국어로 답변해주세요."""
+
+# --- License ---
+LICENSE """Apache-2.0"""

gguf/Modelfile.3b-v2-Q8_0

@@ -0,0 +1,24 @@
+# FRANKENSTALLM 3B v2 Korean ORPO — Ollama Modelfile (Q8_0)
+# Usage:
+#   ollama create frankenstallm-3b-v2:Q8_0 -f Modelfile.3b-v2-Q8
+#   ollama run frankenstallm-3b-v2:Q8_0
+#
+# Sampling config: ORPO eval grid best (t0.7_rep1.2)
+#   3-gram rep=0%, EOS=100%, avg_tokens=189.2
+
+FROM ./outputs/gguf/frankenstallm-3b-v2-Q8_0.gguf
+
+# --- Sampling Config (ORPO eval grid + Ollama 검증) ---
+PARAMETER temperature 0.7
+PARAMETER repeat_penalty 1.2
+PARAMETER top_p 0.9
+PARAMETER top_k 50
+PARAMETER num_predict 512
+PARAMETER num_ctx 4096
+PARAMETER stop "</s>"
+
+# --- System Prompt ---
+SYSTEM """당신은 FRANKENSTALLM, 한국어에 특화된 AI 어시스턴트입니다. 정확하고 자연스러운 한국어로 답변해주세요."""
+
+# --- License ---
+LICENSE """Apache-2.0"""

gguf/Modelfile.3b-v2-f16

@@ -0,0 +1,24 @@
+# FRANKENSTALLM 3B v2 Korean ORPO — Ollama Modelfile (f16, full precision)
+# Usage:
+#   ollama create frankenstallm-3b-v2:f16 -f Modelfile.3b-v2-f16
+#   ollama run frankenstallm-3b-v2:f16
+#
+# Sampling config: ORPO eval grid best (t0.7_rep1.2)
+#   3-gram rep=0%, EOS=100%, avg_tokens=189.2
+
+FROM ./outputs/gguf/frankenstallm-3b-v2-f16.gguf
+
+# --- Sampling Config (ORPO eval grid + Ollama 검증) ---
+PARAMETER temperature 0.7
+PARAMETER repeat_penalty 1.2
+PARAMETER top_p 0.9
+PARAMETER top_k 50
+PARAMETER num_predict 512
+PARAMETER num_ctx 4096
+PARAMETER stop "</s>"
+
+# --- System Prompt ---
+SYSTEM """당신은 FRANKENSTALLM, 한국어에 특화된 AI 어시스턴트입니다. 정확하고 자연스러운 한국어로 답변해주세요."""
+
+# --- License ---
+LICENSE """Apache-2.0"""

gguf/frankenstallm-3b-Q4_K_M.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c5dfff7f79d494f1ed2ccfb262e2ed2fb2ab73f746a7d81bc37cff2913940a7
+size 1961854240

gguf/frankenstallm-3b-Q8_0.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1fbeba7ce0119d511000381708b8267583353de1dab272f286b76361c4eeb826
+size 3414873376

gguf/frankenstallm-3b-f16.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dbd4502f15a565412cbbc54b0b0eff4791a38f7f71c7f91efe89730d670e4e52
+size 6425924896

gguf/frankenstallm-3b-v2-Q4_K_M.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8bbd10a085363be66372f45899cc6625752c4deb3d0f85878e58a3aca34f86f4
+size 1851912864

gguf/frankenstallm-3b-v2-Q8_0.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ae74b6daeb4ac00146a7a34403e584d5fc3913045372eac927bf719f5e18b95
+size 3206818464

gguf/frankenstallm-3b-v2-f16.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be0982394fce045db331d375002c39f5f73978c94fec470a682192ba56014170
+size 6034287264