Spaces:
Running on Zero
Running on Zero
Update app.py
Browse files
app.py
CHANGED
|
@@ -34,6 +34,9 @@ MODEL_IDS: List[str] = [
|
|
| 34 |
"HuggingFaceTB/SmolLM2-135M-Instruct",
|
| 35 |
"OpenCerebral/Boris-1.3-125M-Instruct",
|
| 36 |
"OpenCerebral/Boris-1.3-75M-Instruct",
|
|
|
|
|
|
|
|
|
|
| 37 |
]
|
| 38 |
|
| 39 |
MODEL_DISPLAY: Dict[str, str] = {
|
|
@@ -43,6 +46,9 @@ MODEL_DISPLAY: Dict[str, str] = {
|
|
| 43 |
"HuggingFaceTB/SmolLM2-135M-Instruct": "SmolLM2-135M-Instruct",
|
| 44 |
"OpenCerebral/Boris-1.3-125M-Instruct": "Boris-1.3-125M-Instruct",
|
| 45 |
"OpenCerebral/Boris-1.3-75M-Instruct": "Boris-1.3-75M-Instruct",
|
|
|
|
|
|
|
|
|
|
| 46 |
}
|
| 47 |
|
| 48 |
BASE_MODEL_IDS: List[str] = [
|
|
@@ -56,6 +62,9 @@ BASE_MODEL_IDS: List[str] = [
|
|
| 56 |
"fromziro/Er-Large-30M",
|
| 57 |
"fromziro/Negative-v1.0",
|
| 58 |
"fromziro/Syn-2.6M",
|
|
|
|
|
|
|
|
|
|
| 59 |
]
|
| 60 |
|
| 61 |
BASE_MODEL_DISPLAY: Dict[str, str] = {
|
|
@@ -69,6 +78,9 @@ BASE_MODEL_DISPLAY: Dict[str, str] = {
|
|
| 69 |
"fromziro/Er-Large-30M": "Er-Large-30M",
|
| 70 |
"fromziro/Negative-v1.0": "Negative-v1.0",
|
| 71 |
"fromziro/Syn-2.6M": "Syn-2.6M",
|
|
|
|
|
|
|
|
|
|
| 72 |
}
|
| 73 |
|
| 74 |
MODEL_PARAMS: Dict[str, float] = {
|
|
@@ -88,6 +100,12 @@ MODEL_PARAMS: Dict[str, float] = {
|
|
| 88 |
"fromziro/Er-Large-30M": 31.9e6,
|
| 89 |
"fromziro/Negative-v1.0": 67.7e3,
|
| 90 |
"fromziro/Syn-2.6M": 2.6e6,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 91 |
}
|
| 92 |
|
| 93 |
FALLBACK_IDS: Dict[str, str] = {}
|
|
@@ -174,6 +192,12 @@ GEN_DEFAULTS: Dict[str, dict] = {
|
|
| 174 |
"fromziro/Er-Large-30M": {"max_new_tokens": 64, "temperature": 0.8, "top_p": 0.95, "repetition_penalty": 1.1, "do_sample": True},
|
| 175 |
"fromziro/Negative-v1.0": {"max_new_tokens": 48, "temperature": 0.9, "top_p": 0.95, "repetition_penalty": 1.1, "do_sample": True},
|
| 176 |
"fromziro/Syn-2.6M": {"max_new_tokens": 48, "temperature": 0.8, "top_p": 0.95, "repetition_penalty": 1.1, "do_sample": True},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 177 |
}
|
| 178 |
|
| 179 |
MODEL_CONTEXT: Dict[str, int] = {
|
|
@@ -193,6 +217,12 @@ MODEL_CONTEXT: Dict[str, int] = {
|
|
| 193 |
"fromziro/Er-Large-30M": 2048,
|
| 194 |
"fromziro/Negative-v1.0": 96,
|
| 195 |
"fromziro/Syn-2.6M": 384,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 196 |
}
|
| 197 |
|
| 198 |
|
|
|
|
| 34 |
"HuggingFaceTB/SmolLM2-135M-Instruct",
|
| 35 |
"OpenCerebral/Boris-1.3-125M-Instruct",
|
| 36 |
"OpenCerebral/Boris-1.3-75M-Instruct",
|
| 37 |
+
"BananaMind/BananaMind-2-Pro-Preview-Chat",
|
| 38 |
+
"BananaMind/BananaMind-2-Nano-Chat",
|
| 39 |
+
"BananaMind/BananaMind-2-Mini-Chat",
|
| 40 |
]
|
| 41 |
|
| 42 |
MODEL_DISPLAY: Dict[str, str] = {
|
|
|
|
| 46 |
"HuggingFaceTB/SmolLM2-135M-Instruct": "SmolLM2-135M-Instruct",
|
| 47 |
"OpenCerebral/Boris-1.3-125M-Instruct": "Boris-1.3-125M-Instruct",
|
| 48 |
"OpenCerebral/Boris-1.3-75M-Instruct": "Boris-1.3-75M-Instruct",
|
| 49 |
+
"BananaMind/BananaMind-2-Pro-Preview-Chat": "BananaMind-2-Pro-Preview-Chat",
|
| 50 |
+
"BananaMind/BananaMind-2-Nano-Chat": "BananaMind-2-Nano-Chat",
|
| 51 |
+
"BananaMind/BananaMind-2-Mini-Chat": "BananaMind-2-Mini-Chat",
|
| 52 |
}
|
| 53 |
|
| 54 |
BASE_MODEL_IDS: List[str] = [
|
|
|
|
| 62 |
"fromziro/Er-Large-30M",
|
| 63 |
"fromziro/Negative-v1.0",
|
| 64 |
"fromziro/Syn-2.6M",
|
| 65 |
+
"BananaMind/BananaMind-2-Medium",
|
| 66 |
+
"BananaMind/BananaMind-2-Mini",
|
| 67 |
+
"BananaMind/BananaMind-2-Nano",
|
| 68 |
]
|
| 69 |
|
| 70 |
BASE_MODEL_DISPLAY: Dict[str, str] = {
|
|
|
|
| 78 |
"fromziro/Er-Large-30M": "Er-Large-30M",
|
| 79 |
"fromziro/Negative-v1.0": "Negative-v1.0",
|
| 80 |
"fromziro/Syn-2.6M": "Syn-2.6M",
|
| 81 |
+
"BananaMind/BananaMind-2-Medium": "BananaMind-2-Medium",
|
| 82 |
+
"BananaMind/BananaMind-2-Mini": "BananaMind-2-Mini",
|
| 83 |
+
"BananaMind/BananaMind-2-Nano": "BananaMind-2-Nano",
|
| 84 |
}
|
| 85 |
|
| 86 |
MODEL_PARAMS: Dict[str, float] = {
|
|
|
|
| 100 |
"fromziro/Er-Large-30M": 31.9e6,
|
| 101 |
"fromziro/Negative-v1.0": 67.7e3,
|
| 102 |
"fromziro/Syn-2.6M": 2.6e6,
|
| 103 |
+
"BananaMind/BananaMind-2-Pro-Preview-Chat": 139.0e6,
|
| 104 |
+
"BananaMind/BananaMind-2-Nano-Chat": 9.97e6,
|
| 105 |
+
"BananaMind/BananaMind-2-Mini-Chat": 25.18e6,
|
| 106 |
+
"BananaMind/BananaMind-2-Medium": 55.85e6,
|
| 107 |
+
"BananaMind/BananaMind-2-Mini": 28.32e6,
|
| 108 |
+
"BananaMind/BananaMind-2-Nano": 12.07e6,
|
| 109 |
}
|
| 110 |
|
| 111 |
FALLBACK_IDS: Dict[str, str] = {}
|
|
|
|
| 192 |
"fromziro/Er-Large-30M": {"max_new_tokens": 64, "temperature": 0.8, "top_p": 0.95, "repetition_penalty": 1.1, "do_sample": True},
|
| 193 |
"fromziro/Negative-v1.0": {"max_new_tokens": 48, "temperature": 0.9, "top_p": 0.95, "repetition_penalty": 1.1, "do_sample": True},
|
| 194 |
"fromziro/Syn-2.6M": {"max_new_tokens": 48, "temperature": 0.8, "top_p": 0.95, "repetition_penalty": 1.1, "do_sample": True},
|
| 195 |
+
"BananaMind/BananaMind-2-Pro-Preview-Chat": {"max_new_tokens": 64, "temperature": 0.7, "top_p": 0.9, "repetition_penalty": 1.1, "do_sample": True},
|
| 196 |
+
"BananaMind/BananaMind-2-Nano-Chat": {"max_new_tokens": 64, "temperature": 0.7, "top_p": 0.9, "repetition_penalty": 1.1, "do_sample": True},
|
| 197 |
+
"BananaMind/BananaMind-2-Mini-Chat": {"max_new_tokens": 64, "temperature": 0.7, "top_p": 0.9, "repetition_penalty": 1.1, "do_sample": True},
|
| 198 |
+
"BananaMind/BananaMind-2-Medium": {"max_new_tokens": 64, "temperature": 0.8, "top_p": 0.95, "repetition_penalty": 1.1, "do_sample": True},
|
| 199 |
+
"BananaMind/BananaMind-2-Mini": {"max_new_tokens": 64, "temperature": 0.8, "top_p": 0.95, "repetition_penalty": 1.1, "do_sample": True},
|
| 200 |
+
"BananaMind/BananaMind-2-Nano": {"max_new_tokens": 64, "temperature": 0.8, "top_p": 0.95, "repetition_penalty": 1.1, "do_sample": True},
|
| 201 |
}
|
| 202 |
|
| 203 |
MODEL_CONTEXT: Dict[str, int] = {
|
|
|
|
| 217 |
"fromziro/Er-Large-30M": 2048,
|
| 218 |
"fromziro/Negative-v1.0": 96,
|
| 219 |
"fromziro/Syn-2.6M": 384,
|
| 220 |
+
"BananaMind/BananaMind-2-Pro-Preview-Chat": 3072,
|
| 221 |
+
"BananaMind/BananaMind-2-Nano-Chat": 4096,
|
| 222 |
+
"BananaMind/BananaMind-2-Mini-Chat": 4096,
|
| 223 |
+
"BananaMind/BananaMind-2-Medium": 3072,
|
| 224 |
+
"BananaMind/BananaMind-2-Mini": 4096,
|
| 225 |
+
"BananaMind/BananaMind-2-Nano": 4096,
|
| 226 |
}
|
| 227 |
|
| 228 |
|