Upload HfMoondream

model-00001-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a9414434ab3afb560b37bbd5d3972ae944679e7773a60ece538e4231d2cf142f
+oid sha256:fd4b3d0d6daae9c4212056cd64f02f408ff083bbb0244114eecd05fcba30037e
 size 4907406296

model-00002-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b0433bb359387b93502680ac120913f46e0d6d62940f74ef75759a085edcad86
+oid sha256:7cf6d17391db58801b61173510ba629875679dbcbe4bfd3cb38ac0958b3c70a0
 size 4736548872

model-00003-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:82b15aaadff6efa4013788ccaa321d496993fe41240305c7b8dd8e8cfbc4fa69
+oid sha256:f4391c6d6b46ed49aa00afddf1f7df9dd0845cbc681fdaf424e727b01ea2d3e4
 size 4502742464

model-00004-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fde1839f3766d227b30cfa07521cc0126bfacac08e91f135defdc7624405977f
+oid sha256:6af14858bdd7cdea5d19d786726e48434b02a3c0c52a771a0f25b6a8ca640187
 size 4390620392

moondream.py

@@ -51,7 +51,7 @@ ObjectSamplingSettings = TypedDict(
 DEFAULT_MAX_TOKENS = 768
 DEFAULT_TEMPERATURE = 0.5
 DEFAULT_TOP_P = 0.9
-DEFAULT_MAX_OBJECTS = 50
+DEFAULT_MAX_OBJECTS = 150
 
 
 @dataclass(frozen=True)
@@ -590,10 +590,6 @@ class MoondreamModel(nn.Module):
                     logits_BV, _ = self._decode_one_tok(next_emb, mask, pos_ids, lora)
                     logits_BV[:, self.config.tokenizer.answer_id] = float("-inf")
 
-                    # Suppress EOS for the first token to ensure at least one answer token
-                    if generated_tokens == 0:
-                        logits_BV[:, eos_id] = float("-inf")
-
                     pos += 1
 
                     if temperature == 0:
@@ -620,7 +616,7 @@ class MoondreamModel(nn.Module):
         self,
         image: Optional[Union[Image.Image, EncodedImage]] = None,
         question: str = None,
-        reasoning: bool = False,
+        reasoning: bool = True,
         spatial_refs: Optional[SpatialRefs] = None,
         stream: bool = False,
         settings: Optional[TextSamplingSettings] = None,
@@ -662,10 +658,7 @@ class MoondreamModel(nn.Module):
                     spatial_toks.extend([coord_id, coord_id, size_id])
 
         prompt_tokens = [
-            prompt_toks
-            + spatial_toks
-            + self.tokenizer.encode(question).ids
-            + self.config.tokenizer.templates["query"]["suffix"]
+            prompt_toks + spatial_toks + self.tokenizer.encode(question).ids
         ]
 
         if reasoning: