Merge branch 'master' of https://git.d-popov.com/popov/neuron-tai

packages/node/meshnet_node/model_backend.py

@@ -482,12 +482,36 @@ def _load_partial_model_from_snapshot(
     if callable(tie_weights):
         tie_weights()
 
+    # Multimodal/MTP checkpoints (e.g. Qwen3.5/3.6-MoE) carry vision and
+    # multi-token-prediction tensors the text-only CausalLM never builds;
+    # transformers' from_pretrained drops them via _keys_to_ignore_on_load_unexpected,
+    # so the manual loader must skip them too.
+    expected_keys = _model_state_dict_keys(model)
     tensors_by_file: dict[str, list[str]] = {}
+    skipped: list[str] = []
     for tensor_name in sorted(tensor_names):
         rel_file = weight_map.get(tensor_name)
         if not isinstance(rel_file, str):
             continue
+        if (
+            expected_keys is not None
+            and _checkpoint_tensor_name_for_model(model, tensor_name) not in expected_keys
+        ):
+            skipped.append(tensor_name)
+            continue
         tensors_by_file.setdefault(rel_file, []).append(tensor_name)
+    if skipped:
+        preview = ", ".join(skipped[:3])
+        print(
+            f"  Skipping {len(skipped)} checkpoint tensors absent from the causal LM "
+            f"(e.g. {preview})",
+            flush=True,
+        )
+    if not tensors_by_file:
+        raise PartialModelLoadUnsupported(
+            f"no checkpoint tensors for layers {shard_start}-{shard_end} match the "
+            f"causal LM built from {snapshot_dir}"
+        )
 
     for rel_file, names in tensors_by_file.items():
         checkpoint_file = snapshot_dir / rel_file
@@ -584,6 +608,17 @@ def _causal_lm_config(cfg: Any) -> Any:
     return cfg
 
 
+def _model_state_dict_keys(model: Any) -> set[str] | None:
+    """Expected parameter/buffer names, or None when the model can't report them."""
+    state_dict = getattr(model, "state_dict", None)
+    if not callable(state_dict):
+        return None
+    try:
+        return set(state_dict().keys())
+    except Exception:
+        return None
+
+
 def _checkpoint_tensor_name_for_model(model: Any, tensor_name: str) -> str:
     """Map multimodal checkpoint keys onto text-only CausalLM modules when needed."""
     inner = getattr(model, "model", None)

packages/tracker/meshnet_tracker/dashboard.html

@@ -5,8 +5,9 @@
 <meta name="viewport" content="width=device-width, initial-scale=1">
 <title>meshnet tracker</title>
 <style>
-  :root { --bg:#0d1117; --panel:#161b22; --border:#30363d; --fg:#c9d1d9;
-          --dim:#8b949e; --accent:#58a6ff; --ok:#3fb950; --bad:#f85149; --warn:#d29922; }
+  :root { --bg:#0d1117; --panel:#161b22; --border:#30363d; --fg:#e6edf3;
+          --dim:#8b949e; --accent:#58a6ff; --ok:#3fb950; --bad:#f85149; --warn:#d29922;
+          --chat-input-bg:#21262d; --chat-user-bg:#1f4788; --chat-user-border:#388bfd; }
   * { box-sizing:border-box; }
   html, body { height:100%; }
   body { margin:0; background:var(--bg); color:var(--fg);
@@ -114,7 +115,7 @@
   .chat-session-item:hover .chat-session-delete,
   .chat-session-item.active .chat-session-delete { opacity:1; }
   .chat-session-delete:hover { color:var(--bad); background:#1a1012; }
-  .chat-main { display:flex; flex-direction:column; min-height:0; min-width:0; }
+  .chat-main { display:flex; flex-direction:column; min-height:0; min-width:0; color-scheme:dark; }
   .chat-toolbar {
     display:flex; gap:12px; align-items:center; flex-shrink:0;
     padding:10px 16px; border-bottom:1px solid var(--border); background:var(--panel);
@@ -122,10 +123,15 @@
   .chat-toolbar label {
     display:flex; align-items:center; gap:8px; color:var(--dim); margin:0;
   }
-  .chat-toolbar select { min-width:220px; max-width:min(420px, 50vw); }
+  .chat-toolbar select {
+    min-width:220px; max-width:min(420px, 50vw);
+    color:var(--fg); background:var(--chat-input-bg); border:1px solid var(--border);
+    border-radius:6px; padding:6px 8px;
+  }
   .chat-status { color:var(--dim); font-size:12px; margin-left:auto; }
   .chat-messages {
     flex:1; overflow:auto; padding:24px 16px; min-height:0;
+    background:var(--bg); color:var(--fg);
   }
   .chat-messages-inner {
     max-width:768px; margin:0 auto; display:flex; flex-direction:column; gap:20px;
@@ -139,13 +145,15 @@
   .chat-row.assistant, .chat-row.error { justify-content:flex-start; }
   .chat-bubble {
     max-width:85%; padding:12px 14px; border-radius:16px; line-height:1.55;
-    white-space:pre-wrap; word-break:break-word; font-size:13px;
+    white-space:pre-wrap; word-break:break-word; font-size:14px; color:var(--fg);
   }
   .chat-bubble.user {
-    background:#1f3a5f; border:1px solid #264a73; border-bottom-right-radius:4px;
+    background:var(--chat-user-bg); border:1px solid var(--chat-user-border);
+    border-bottom-right-radius:4px; color:#f0f6fc;
   }
   .chat-bubble.assistant {
-    background:transparent; border:0; padding-left:0; padding-right:0; max-width:100%;
+    background:var(--panel); border:1px solid var(--border);
+    border-bottom-left-radius:4px; max-width:100%; color:var(--fg);
   }
   .chat-bubble.error {
     background:#1a1012; border:1px solid #5c2020; color:#ffb4b4; border-bottom-left-radius:4px;
@@ -157,16 +165,25 @@
   .chat-compose {
     display:flex; gap:8px; align-items:flex-end; max-width:768px; margin:0 auto;
     padding:10px 12px; border:1px solid var(--border); border-radius:16px;
-    background:var(--bg);
+    background:var(--chat-input-bg);
+  }
+  .chat-compose:focus-within {
+    border-color:var(--accent);
+    box-shadow:0 0 0 1px var(--accent);
   }
-  .chat-compose:focus-within { border-color:var(--accent); }
   .chat-compose textarea {
     flex:1; min-height:24px; max-height:200px; resize:none; width:auto;
-    border:0; background:transparent; padding:2px 0; outline:none;
+    border:0; background:transparent; padding:4px 0; outline:none;
+    color:var(--fg); caret-color:var(--accent); font:inherit; font-size:14px; line-height:1.5;
   }
+  .chat-compose textarea::placeholder { color:var(--dim); opacity:1; }
   .chat-compose button {
     flex-shrink:0; min-width:36px; height:36px; padding:0;
-    border-radius:8px; border:1px solid var(--border);
+    border-radius:8px; border:1px solid var(--chat-user-border);
+    background:var(--chat-user-bg); color:#f0f6fc;
+  }
+  .chat-compose button:hover:not(:disabled) {
+    border-color:var(--accent); background:#2563b8;
   }
   .chat-compose button:disabled { opacity:.45; cursor:not-allowed; }
   .console {

tests/test_real_model_backend.py

@@ -559,6 +559,96 @@ def test_partial_snapshot_loader_remaps_language_model_checkpoint_keys(tmp_path)
     assert set_calls == ["model.layers.1.self_attn.q_proj.weight"]
 
 
+def test_partial_snapshot_loader_skips_tensors_absent_from_causal_lm(tmp_path):
+    # Multimodal/MTP checkpoints (Qwen3.5/3.6-MoE) carry mtp.* and model.visual.*
+    # tensors that the text-only CausalLM never builds — they must be skipped,
+    # not assigned (assignment raises AttributeError: 'mtp' / 'visual').
+    snapshot_dir = tmp_path / "snapshot"
+    snapshot_dir.mkdir()
+    (snapshot_dir / "config.json").write_text(json.dumps({
+        "text_config": {"num_hidden_layers": 3},
+    }))
+    (snapshot_dir / "model.safetensors.index.json").write_text(json.dumps({
+        "weight_map": {
+            "model.language_model.layers.1.self_attn.q_proj.weight": "shard-2.safetensors",
+            "mtp.layers.1.input_layernorm.weight": "shard-2.safetensors",
+            "model.visual.blocks.1.attn.qkv.weight": "shard-2.safetensors",
+        }
+    }))
+    (snapshot_dir / "shard-2.safetensors").write_bytes(b"stub")
+
+    class FakeModule:
+        def to(self, device):
+            return self
+
+    class FakeModel:
+        def __init__(self):
+            self.model = types.SimpleNamespace(
+                layers=[FakeModule(), FakeModule(), FakeModule()],
+                rotary_emb=FakeModule(),
+            )
+
+        def tie_weights(self):
+            pass
+
+        def state_dict(self):
+            return {"model.layers.1.self_attn.q_proj.weight": None}
+
+    class AutoConfigStub:
+        @staticmethod
+        def from_pretrained(model_id):
+            return types.SimpleNamespace(
+                text_config=types.SimpleNamespace(num_hidden_layers=3),
+                get_text_config=lambda: types.SimpleNamespace(num_hidden_layers=3),
+            )
+
+    class AutoModelStub:
+        @staticmethod
+        def from_config(cfg, torch_dtype=None):
+            return FakeModel()
+
+    set_calls = []
+
+    def fake_set_tensor(module, tensor_name, device, value=None, dtype=None):
+        set_calls.append(tensor_name)
+
+    class FakeSafeOpen:
+        def __init__(self, filename, framework, device):
+            pass
+
+        def __enter__(self):
+            return self
+
+        def __exit__(self, exc_type, exc, tb):
+            return False
+
+        def get_tensor(self, tensor_name):
+            return tensor_name
+
+    class UnusedContext:
+        def __enter__(self):
+            return None
+
+        def __exit__(self, exc_type, exc, tb):
+            return False
+
+    _load_partial_model_from_snapshot(
+        AutoConfigStub,
+        AutoModelStub,
+        types.SimpleNamespace(),
+        str(snapshot_dir),
+        1,
+        1,
+        "bf16",
+        "cpu:0",
+        init_empty_weights_fn=UnusedContext,
+        set_tensor_fn=fake_set_tensor,
+        safe_open_fn=FakeSafeOpen,
+    )
+
+    assert set_calls == ["model.layers.1.self_attn.q_proj.weight"]
+
+
 def test_partial_snapshot_loader_materializes_only_assigned_tensors(tmp_path):
     snapshot_dir = tmp_path / "snapshot"
     snapshot_dir.mkdir()