Update README.md

README.md

@@ -86,8 +86,9 @@ python -m sglang.launch_server \
     --pp-size 1 \
     --dp-size 1 \
     --trust-remote-code \
+    --context-length 262144 \
     --tool-call-parser qwen25 \
-    --json-model-override-args '{"linear_backend": "seg_la", "torch_dtype": "bfloat16", "architectures": ["BailingMoeV2_5ForCausalLM"], "model_type": "bailing_hybrid", "rope_scaling": {"rope_type": "yarn", "factor": 2.0, "rope_theta": 6000000, "partial_rotary_factor": 0.5, "original_max_position_embeddings": 131072}}' \
+    --json-model-override-args '{"rope_scaling": {"rope_type": "yarn", "factor": 2.0, "rope_theta": 6000000, "partial_rotary_factor": 0.5, "original_max_position_embeddings": 131072}}' \
     --dist-init-addr $MASTER_IP:2345 \
     --port $PORT \
     --nnodes 1
@@ -111,6 +112,8 @@ python -m sglang.launch_server \
     --tp-size 4 \
     --pp-size 1 \
     --dp-size 1 \
+    --context-length 262144 \
+    --mamba-scheduler-strategy extra_buffer \
     --speculative-algorithm NEXTN \
     --speculative-num-steps 3 \
     --speculative-eagle-topk 1 \
@@ -119,7 +122,7 @@ python -m sglang.launch_server \
     --max-running-requests 64 \
     --max-mamba-cache-size 256 \
     --tool-call-parser qwen25 \
-    --json-model-override-args '{"linear_backend": "seg_la", "torch_dtype": "bfloat16", "architectures": ["BailingMoeV2_5ForCausalLM"], "model_type": "bailing_hybrid", "rope_scaling": {"rope_type": "yarn", "factor": 2.0, "rope_theta": 6000000, "partial_rotary_factor": 0.5, "original_max_position_embeddings": 131072}}' \
+    --json-model-override-args '{"rope_scaling": {"rope_type": "yarn", "factor": 2.0, "rope_theta": 6000000, "partial_rotary_factor": 0.5, "original_max_position_embeddings": 131072}}' \
     --trust-remote-code \
     --dist-init-addr $MASTER_IP:2345 \
     --port $PORT \