diff --git a/benchmarks/multi_node/srt-slurm-recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p1d-tp4-dep4-eagle3-8k1k.yaml b/benchmarks/multi_node/srt-slurm-recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p1d-tp4-dep4-eagle3-8k1k.yaml
new file mode 100644
index 000000000..a655ade33
--- /dev/null
+++ b/benchmarks/multi_node/srt-slurm-recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p1d-tp4-dep4-eagle3-8k1k.yaml
@@ -0,0 +1,98 @@
+name: "minimax-m3-vllm-disagg-b200-1p1d-tp4-dep4-fp4-8k1k-eagle3"
+
+model:
+  path: "nvidia/MiniMax-M3-NVFP4"
+  container: "vllm/vllm-openai:nightly-5e35a6f4f9bbc217c599692157ca985c894373f7"
+  precision: fp4
+
+resources:
+  gpu_type: b200
+  gpus_per_node: 8
+  prefill_nodes: 1
+  decode_nodes: 0
+  prefill_workers: 1
+  decode_workers: 1
+  gpus_per_prefill: 4
+  gpus_per_decode: 4
+
+dynamo:
+  install: true
+  version: 1.3.0.dev20260710
+
+frontend:
+  type: dynamo
+  enable_multiple_frontends: false
+
+backend:
+  type: vllm
+  connector: null
+  allow_prefill_decode_colocation: true
+  allow_prefill_decode_colocation_across_nodes: true
+
+  prefill_environment:
+    VLLM_FLOAT32_MATMUL_PRECISION: high
+    VLLM_FLASHINFER_ALLREDUCE_BACKEND: trtllm
+    UCX_TLS: cuda_copy,cuda_ipc,rc
+
+  decode_environment:
+    VLLM_FLOAT32_MATMUL_PRECISION: high
+    VLLM_FLASHINFER_ALLREDUCE_BACKEND: trtllm
+    VLLM_MINIMAX_M3_MSA_DECODE_BACKEND: cutlass
+    UCX_TLS: cuda_copy,cuda_ipc,rc
+    VLLM_MINIMAX_M3_MSA_DECODE_BACKEND: cutlass
+
+  vllm_config:
+    prefill:
+      no-enable-flashinfer-autotune: true
+      tensor-parallel-size: 4
+      pipeline-parallel-size: 1
+      trust-remote-code: true
+      no-enable-prefix-caching: true
+      kv-transfer-config: '{"kv_connector": "NixlConnector", "kv_role": "kv_both"}'
+      attention-config: '{"backend": "FLASHINFER", "use_trtllm_attention": true, "indexer_kv_dtype": "fp8"}'
+      block-size: 128
+      gpu-memory-utilization: 0.9
+      max-model-len: 9472
+      language-model-only: true
+      kv-cache-dtype: fp8
+      speculative-config: '{"method":"eagle3","model":"Inferact/MiniMax-M3-EAGLE3-GQA","num_speculative_tokens":1,"attention_backend":"FLASH_ATTN"}'
+      stream-interval: 32
+      max-cudagraph-capture-size: 2048
+      max-num-batched-tokens: 16384
+
+    decode:
+      no-enable-flashinfer-autotune: true
+      tensor-parallel-size: 1
+      pipeline-parallel-size: 1
+      data-parallel-size: 4
+      data-parallel-rpc-port: 13345
+      enable-expert-parallel: true
+      trust-remote-code: true
+      no-enable-prefix-caching: true
+      kv-transfer-config: '{"kv_connector": "NixlConnector", "kv_role": "kv_both"}'
+      attention-config: '{"backend": "FLASHINFER", "use_trtllm_attention": true, "indexer_kv_dtype": "fp8", "minimax_m3_msa_decode_backend": "cutlass"}'
+      block-size: 128
+      gpu-memory-utilization: 0.9
+      max-model-len: 9472
+      language-model-only: true
+      kv-cache-dtype: fp8
+      speculative-config: '{"method":"eagle3","model":"Inferact/MiniMax-M3-EAGLE3-GQA","num_speculative_tokens":3,"attention_backend":"FLASH_ATTN"}'
+      stream-interval: 32
+      max-num-seqs: 1024
+      max-num-batched-tokens: 16384
+      max-cudagraph-capture-size: 2048
+      compilation-config: '{"cudagraph_mode":"FULL_DECODE_ONLY"}'
+
+health_check:
+  max_attempts: 360
+  interval_seconds: 10
+
+benchmark:
+  type: sa-bench
+  isl: 8192
+  osl: 1024
+  req_rate: inf
+  num_warmup_mult: 0
+  random_range_ratio: 0.8
+  use_chat_template: true
+  concurrencies: "256x512"
diff --git a/benchmarks/multi_node/srt-slurm-recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p1d-tp4-tp4-eagle3-8k1k.yaml b/benchmarks/multi_node/srt-slurm-recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p1d-tp4-tp4-eagle3-8k1k.yaml
new file mode 100644
index 000000000..ff498c89e
--- /dev/null
+++ b/benchmarks/multi_node/srt-slurm-recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p1d-tp4-tp4-eagle3-8k1k.yaml
@@ -0,0 +1,91 @@
+name: "minimax-m3-vllm-disagg-b200-1p1d-tp4-tp4-fp4-8k1k-eagle3"
+
+model:
+  path: "nvidia/MiniMax-M3-NVFP4"
+  container: "vllm/vllm-openai:nightly-5e35a6f4f9bbc217c599692157ca985c894373f7"
+  precision: fp4
+
+resources:
+  gpu_type: b200
+  gpus_per_node: 8
+  prefill_nodes: 1
+  decode_nodes: 0
+  prefill_workers: 1
+  decode_workers: 1
+  gpus_per_prefill: 4
+  gpus_per_decode: 4
+
+dynamo:
+  install: true
+  version: 1.3.0.dev20260710
+
+frontend:
+  type: dynamo
+  enable_multiple_frontends: false
+
+backend:
+  type: vllm
+  connector: null
+  allow_prefill_decode_colocation: true
+  allow_prefill_decode_colocation_across_nodes: true
+
+  prefill_environment:
+    VLLM_FLOAT32_MATMUL_PRECISION: high
+    UCX_TLS: cuda_copy,cuda_ipc,rc
+
+  decode_environment:
+    VLLM_FLOAT32_MATMUL_PRECISION: high
+    UCX_TLS: cuda_copy,cuda_ipc,rc
+    VLLM_MINIMAX_M3_MSA_DECODE_BACKEND: cutlass
+
+  vllm_config:
+    prefill:
+      no-enable-flashinfer-autotune: true
+      tensor-parallel-size: 4
+      pipeline-parallel-size: 1
+      trust-remote-code: true
+      no-enable-prefix-caching: true
+      kv-transfer-config: '{"kv_connector": "NixlConnector", "kv_role": "kv_both"}'
+      attention-config: '{"backend": "FLASHINFER", "use_trtllm_attention": true, "indexer_kv_dtype": "fp8"}'
+      block-size: 128
+      gpu-memory-utilization: 0.9
+      max-model-len: 9472
+      language-model-only: true
+      speculative-config: '{"method":"eagle3","model":"Inferact/MiniMax-M3-EAGLE3-GQA","num_speculative_tokens":1,"attention_backend":"FLASH_ATTN"}'
+      stream-interval: 32
+      max-cudagraph-capture-size: 2048
+      max-num-batched-tokens: 16384
+
+    decode:
+      no-enable-flashinfer-autotune: true
+      tensor-parallel-size: 4
+      pipeline-parallel-size: 1
+      enable-expert-parallel: false
+      trust-remote-code: true
+      no-enable-prefix-caching: true
+      kv-transfer-config: '{"kv_connector": "NixlConnector", "kv_role": "kv_both"}'
+      attention-config: '{"backend": "FLASHINFER", "use_trtllm_attention": true, "indexer_kv_dtype": "fp8"}'
+      block-size: 128
+      gpu-memory-utilization: 0.9
+      max-model-len: 9472
+      language-model-only: true
+      speculative-config: '{"method":"eagle3","model":"Inferact/MiniMax-M3-EAGLE3-GQA","num_speculative_tokens":3,"attention_backend":"FLASH_ATTN"}'
+      stream-interval: 32
+      max-num-seqs: 1024
+      max-num-batched-tokens: 16384
+      max-cudagraph-capture-size: 2048
+      compilation-config: '{"cudagraph_mode":"FULL_DECODE_ONLY"}'
+
+health_check:
+  max_attempts: 360
+  interval_seconds: 10
+
+benchmark:
+  type: sa-bench
+  isl: 8192
+  osl: 1024
+  req_rate: inf
+  num_warmup_mult: 0
+  random_range_ratio: 0.8
+  use_chat_template: true
+  concurrencies: "192x256x512"
diff --git a/benchmarks/multi_node/srt-slurm-recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p2d-tp4-tp4-eagle3-8k1k.yaml b/benchmarks/multi_node/srt-slurm-recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p2d-tp4-tp4-eagle3-8k1k.yaml
new file mode 100644
index 000000000..a7909a5fb
--- /dev/null
+++ b/benchmarks/multi_node/srt-slurm-recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p2d-tp4-tp4-eagle3-8k1k.yaml
@@ -0,0 +1,91 @@
+name: "minimax-m3-vllm-disagg-b200-1p2d-tp4-tp4-fp4-8k1k-eagle3"
+
+model:
+  path: "nvidia/MiniMax-M3-NVFP4"
+  container: "vllm/vllm-openai:nightly-5e35a6f4f9bbc217c599692157ca985c894373f7"
+  precision: fp4
+
+resources:
+  gpu_type: b200
+  gpus_per_node: 8
+  prefill_nodes: 1
+  decode_nodes: 1
+  prefill_workers: 1
+  decode_workers: 2
+  gpus_per_prefill: 4
+  gpus_per_decode: 4
+
+dynamo:
+  install: true
+  version: 1.3.0.dev20260710
+
+frontend:
+  type: dynamo
+  enable_multiple_frontends: false
+
+backend:
+  type: vllm
+  connector: null
+  allow_prefill_decode_colocation: true
+  allow_prefill_decode_colocation_across_nodes: true
+
+  prefill_environment:
+    VLLM_FLOAT32_MATMUL_PRECISION: high
+    UCX_TLS: cuda_copy,cuda_ipc,rc
+
+  decode_environment:
+    VLLM_FLOAT32_MATMUL_PRECISION: high
+    UCX_TLS: cuda_copy,cuda_ipc,rc
+    VLLM_MINIMAX_M3_MSA_DECODE_BACKEND: cutlass
+
+  vllm_config:
+    prefill:
+      no-enable-flashinfer-autotune: true
+      tensor-parallel-size: 4
+      pipeline-parallel-size: 1
+      trust-remote-code: true
+      no-enable-prefix-caching: true
+      kv-transfer-config: '{"kv_connector": "NixlConnector", "kv_role": "kv_both"}'
+      attention-config: '{"backend": "FLASHINFER", "use_trtllm_attention": true, "indexer_kv_dtype": "fp8"}'
+      block-size: 128
+      gpu-memory-utilization: 0.9
+      max-model-len: 9472
+      language-model-only: true
+      speculative-config: '{"method":"eagle3","model":"Inferact/MiniMax-M3-EAGLE3-GQA","num_speculative_tokens":1,"attention_backend":"FLASH_ATTN"}'
+      stream-interval: 32
+      max-cudagraph-capture-size: 2048
+      max-num-batched-tokens: 16384
+
+    decode:
+      no-enable-flashinfer-autotune: true
+      tensor-parallel-size: 4
+      pipeline-parallel-size: 1
+      enable-expert-parallel: false
+      trust-remote-code: true
+      no-enable-prefix-caching: true
+      kv-transfer-config: '{"kv_connector": "NixlConnector", "kv_role": "kv_both"}'
+      attention-config: '{"backend": "FLASHINFER", "use_trtllm_attention": true, "indexer_kv_dtype": "fp8"}'
+      block-size: 128
+      gpu-memory-utilization: 0.9
+      max-model-len: 9472
+      language-model-only: true
+      speculative-config: '{"method":"eagle3","model":"Inferact/MiniMax-M3-EAGLE3-GQA","num_speculative_tokens":3,"attention_backend":"FLASH_ATTN"}'
+      stream-interval: 32
+      max-num-seqs: 1024
+      max-num-batched-tokens: 16384
+      max-cudagraph-capture-size: 2048
+      compilation-config: '{"cudagraph_mode":"FULL_DECODE_ONLY"}'
+
+health_check:
+  max_attempts: 360
+  interval_seconds: 10
+
+benchmark:
+  type: sa-bench
+  isl: 8192
+  osl: 1024
+  req_rate: inf
+  num_warmup_mult: 0
+  random_range_ratio: 0.8
+  use_chat_template: true
+  concurrencies: "32x64x96"
diff --git a/benchmarks/multi_node/srt-slurm-recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p3d-tp4-tp4-eagle3-8k1k.yaml b/benchmarks/multi_node/srt-slurm-recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p3d-tp4-tp4-eagle3-8k1k.yaml
new file mode 100644
index 000000000..125b2ed52
--- /dev/null
+++ b/benchmarks/multi_node/srt-slurm-recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p3d-tp4-tp4-eagle3-8k1k.yaml
@@ -0,0 +1,91 @@
+name: "minimax-m3-vllm-disagg-b200-1p3d-tp4-tp4-fp4-8k1k-eagle3"
+
+model:
+  path: "nvidia/MiniMax-M3-NVFP4"
+  container: "vllm/vllm-openai:nightly-5e35a6f4f9bbc217c599692157ca985c894373f7"
+  precision: fp4
+
+resources:
+  gpu_type: b200
+  gpus_per_node: 8
+  prefill_nodes: 1
+  decode_nodes: 1
+  prefill_workers: 1
+  decode_workers: 3
+  gpus_per_prefill: 4
+  gpus_per_decode: 4
+
+dynamo:
+  install: true
+  version: 1.3.0.dev20260710
+
+frontend:
+  type: dynamo
+  enable_multiple_frontends: false
+
+backend:
+  type: vllm
+  connector: null
+  allow_prefill_decode_colocation: true
+  allow_prefill_decode_colocation_across_nodes: true
+
+  prefill_environment:
+    VLLM_FLOAT32_MATMUL_PRECISION: high
+    UCX_TLS: cuda_copy,cuda_ipc,rc
+
+  decode_environment:
+    VLLM_FLOAT32_MATMUL_PRECISION: high
+    UCX_TLS: cuda_copy,cuda_ipc,rc
+    VLLM_MINIMAX_M3_MSA_DECODE_BACKEND: cutlass
+
+  vllm_config:
+    prefill:
+      no-enable-flashinfer-autotune: true
+      tensor-parallel-size: 4
+      pipeline-parallel-size: 1
+      trust-remote-code: true
+      no-enable-prefix-caching: true
+      kv-transfer-config: '{"kv_connector": "NixlConnector", "kv_role": "kv_both"}'
+      attention-config: '{"backend": "FLASHINFER", "use_trtllm_attention": true, "indexer_kv_dtype": "fp8"}'
+      block-size: 128
+      gpu-memory-utilization: 0.9
+      max-model-len: 9472
+      language-model-only: true
+      speculative-config: '{"method":"eagle3","model":"Inferact/MiniMax-M3-EAGLE3-GQA","num_speculative_tokens":1,"attention_backend":"FLASH_ATTN"}'
+      stream-interval: 32
+      max-cudagraph-capture-size: 2048
+      max-num-batched-tokens: 16384
+
+    decode:
+      no-enable-flashinfer-autotune: true
+      tensor-parallel-size: 4
+      pipeline-parallel-size: 1
+      enable-expert-parallel: false
+      trust-remote-code: true
+      no-enable-prefix-caching: true
+      kv-transfer-config: '{"kv_connector": "NixlConnector", "kv_role": "kv_both"}'
+      attention-config: '{"backend": "FLASHINFER", "use_trtllm_attention": true, "indexer_kv_dtype": "fp8"}'
+      block-size: 128
+      gpu-memory-utilization: 0.9
+      max-model-len: 9472
+      language-model-only: true
+      speculative-config: '{"method":"eagle3","model":"Inferact/MiniMax-M3-EAGLE3-GQA","num_speculative_tokens":3,"attention_backend":"FLASH_ATTN"}'
+      stream-interval: 32
+      max-num-seqs: 1024
+      max-num-batched-tokens: 16384
+      max-cudagraph-capture-size: 2048
+      compilation-config: '{"cudagraph_mode":"FULL_DECODE_ONLY"}'
+
+health_check:
+  max_attempts: 360
+  interval_seconds: 10
+
+benchmark:
+  type: sa-bench
+  isl: 8192
+  osl: 1024
+  req_rate: inf
+  num_warmup_mult: 0
+  random_range_ratio: 0.8
+  use_chat_template: true
+  concurrencies: "192"
diff --git a/benchmarks/multi_node/srt-slurm-recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p4d-tp4-tp4-eagle3-8k1k.yaml b/benchmarks/multi_node/srt-slurm-recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p4d-tp4-tp4-eagle3-8k1k.yaml
new file mode 100644
index 000000000..3c867e376
--- /dev/null
+++ b/benchmarks/multi_node/srt-slurm-recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p4d-tp4-tp4-eagle3-8k1k.yaml
@@ -0,0 +1,89 @@
+name: "minimax-m3-vllm-disagg-b200-1p4d-tp4-tp4-fp4-8k1k-eagle3"
+
+model:
+  path: "nvidia/MiniMax-M3-NVFP4"
+  container: "vllm/vllm-openai:nightly-5e35a6f4f9bbc217c599692157ca985c894373f7"
+  precision: fp4
+
+resources:
+  gpu_type: b200
+  gpus_per_node: 8
+  prefill_nodes: 1
+  decode_nodes: 2
+  prefill_workers: 1
+  decode_workers: 4
+  gpus_per_prefill: 4
+  gpus_per_decode: 4
+
+dynamo:
+  install: true
+  version: 1.3.0.dev20260710
+
+frontend:
+  type: dynamo
+  enable_multiple_frontends: false
+
+backend:
+  type: vllm
+  connector: null
+  allow_prefill_decode_colocation: true
+  allow_prefill_decode_colocation_across_nodes: true
+
+  prefill_environment:
+    VLLM_FLOAT32_MATMUL_PRECISION: high
+    UCX_TLS: cuda_copy,cuda_ipc,rc
+
+  decode_environment:
+    VLLM_FLOAT32_MATMUL_PRECISION: high
+    UCX_TLS: cuda_copy,cuda_ipc,rc
+
+  vllm_config:
+    prefill:
+      no-enable-flashinfer-autotune: true
+      tensor-parallel-size: 4
+      pipeline-parallel-size: 1
+      trust-remote-code: true
+      no-enable-prefix-caching: true
+      kv-transfer-config: '{"kv_connector": "NixlConnector", "kv_role": "kv_both"}'
+      attention-config: '{"backend": "FLASHINFER", "use_trtllm_attention": true, "indexer_kv_dtype": "fp8"}'
+      block-size: 128
+      gpu-memory-utilization: 0.9
+      max-model-len: 9472
+      language-model-only: true
+      speculative-config: '{"method":"eagle3","model":"Inferact/MiniMax-M3-EAGLE3-GQA","num_speculative_tokens":1,"attention_backend":"FLASH_ATTN"}'
+      stream-interval: 32
+      max-cudagraph-capture-size: 2048
+      max-num-batched-tokens: 16384
+
+    decode:
+      no-enable-flashinfer-autotune: true
+      tensor-parallel-size: 4
+      pipeline-parallel-size: 1
+      enable-expert-parallel: false
+      trust-remote-code: true
+      no-enable-prefix-caching: true
+      kv-transfer-config: '{"kv_connector": "NixlConnector", "kv_role": "kv_both"}'
+      attention-config: '{"backend": "FLASHINFER", "use_trtllm_attention": true, "indexer_kv_dtype": "fp8"}'
+      block-size: 128
+      gpu-memory-utilization: 0.9
+      max-model-len: 9472
+      language-model-only: true
+      speculative-config: '{"method":"eagle3","model":"Inferact/MiniMax-M3-EAGLE3-GQA","num_speculative_tokens":3,"attention_backend":"FLASH_ATTN"}'
+      stream-interval: 32
+      max-num-seqs: 1024
+      max-num-batched-tokens: 16384
+      max-cudagraph-capture-size: 2048
+
+health_check:
+  max_attempts: 360
+  interval_seconds: 10
+
+benchmark:
+  type: sa-bench
+  isl: 8192
+  osl: 1024
+  req_rate: inf
+  num_warmup_mult: 0
+  random_range_ratio: 0.8
+  use_chat_template: true
+  concurrencies: "4x8x16x32x64"
diff --git a/benchmarks/multi_node/srt-slurm-recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p6d-tp4-tp4-eagle3-8k1k.yaml b/benchmarks/multi_node/srt-slurm-recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p6d-tp4-tp4-eagle3-8k1k.yaml
new file mode 100644
index 000000000..792df6db7
--- /dev/null
+++ b/benchmarks/multi_node/srt-slurm-recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p6d-tp4-tp4-eagle3-8k1k.yaml
@@ -0,0 +1,91 @@
+name: "minimax-m3-vllm-disagg-b200-1p6d-tp4-tp4-fp4-8k1k-eagle3"
+
+model:
+  path: "nvidia/MiniMax-M3-NVFP4"
+  container: "vllm/vllm-openai:nightly-5e35a6f4f9bbc217c599692157ca985c894373f7"
+  precision: fp4
+
+resources:
+  gpu_type: b200
+  gpus_per_node: 8
+  prefill_nodes: 1
+  decode_nodes: 3
+  prefill_workers: 1
+  decode_workers: 6
+  gpus_per_prefill: 4
+  gpus_per_decode: 4
+
+dynamo:
+  install: true
+  version: 1.3.0.dev20260710
+
+frontend:
+  type: dynamo
+  enable_multiple_frontends: false
+
+backend:
+  type: vllm
+  connector: null
+  allow_prefill_decode_colocation: true
+  allow_prefill_decode_colocation_across_nodes: true
+
+  prefill_environment:
+    VLLM_FLOAT32_MATMUL_PRECISION: high
+    UCX_TLS: cuda_copy,cuda_ipc,rc
+
+  decode_environment:
+    VLLM_FLOAT32_MATMUL_PRECISION: high
+    UCX_TLS: cuda_copy,cuda_ipc,rc
+    VLLM_MINIMAX_M3_MSA_DECODE_BACKEND: cutlass
+
+  vllm_config:
+    prefill:
+      no-enable-flashinfer-autotune: true
+      tensor-parallel-size: 4
+      pipeline-parallel-size: 1
+      trust-remote-code: true
+      no-enable-prefix-caching: true
+      kv-transfer-config: '{"kv_connector": "NixlConnector", "kv_role": "kv_both"}'
+      attention-config: '{"backend": "FLASHINFER", "use_trtllm_attention": true, "indexer_kv_dtype": "fp8"}'
+      block-size: 128
+      gpu-memory-utilization: 0.9
+      max-model-len: 9472
+      language-model-only: true
+      speculative-config: '{"method":"eagle3","model":"Inferact/MiniMax-M3-EAGLE3-GQA","num_speculative_tokens":1,"attention_backend":"FLASH_ATTN"}'
+      stream-interval: 32
+      max-cudagraph-capture-size: 2048
+      max-num-batched-tokens: 16384
+
+    decode:
+      no-enable-flashinfer-autotune: true
+      tensor-parallel-size: 4
+      pipeline-parallel-size: 1
+      enable-expert-parallel: false
+      trust-remote-code: true
+      no-enable-prefix-caching: true
+      kv-transfer-config: '{"kv_connector": "NixlConnector", "kv_role": "kv_both"}'
+      attention-config: '{"backend": "FLASHINFER", "use_trtllm_attention": true, "indexer_kv_dtype": "fp8"}'
+      block-size: 128
+      gpu-memory-utilization: 0.9
+      max-model-len: 9472
+      language-model-only: true
+      speculative-config: '{"method":"eagle3","model":"Inferact/MiniMax-M3-EAGLE3-GQA","num_speculative_tokens":3,"attention_backend":"FLASH_ATTN"}'
+      stream-interval: 32
+      max-num-seqs: 1024
+      max-num-batched-tokens: 16384
+      max-cudagraph-capture-size: 2048
+      compilation-config: '{"cudagraph_mode":"FULL_DECODE_ONLY"}'
+
+health_check:
+  max_attempts: 360
+  interval_seconds: 10
+
+benchmark:
+  type: sa-bench
+  isl: 8192
+  osl: 1024
+  req_rate: inf
+  num_warmup_mult: 0
+  random_range_ratio: 0.8
+  use_chat_template: true
+  concurrencies: "12x24x48"
diff --git a/configs/nvidia-master.yaml b/configs/nvidia-master.yaml
index b850f7d6e..39087de3a 100644
--- a/configs/nvidia-master.yaml
+++ b/configs/nvidia-master.yaml
@@ -7783,6 +7783,110 @@ minimaxm3-fp4-b300-dynamo-vllm-8k1k-legacy-max-tput:
           ep: 4
           dp-attn: false
 
+minimaxm3-fp4-b200-dynamo-vllm-mtp:
+  image: vllm/vllm-openai:nightly-5e35a6f4f9bbc217c599692157ca985c894373f7
+  model: nvidia/MiniMax-M3-NVFP4
+  model-prefix: minimaxm3
+  runner: b200-multinode
+  precision: fp4
+  framework: dynamo-vllm
+  router: { name: dynamo-router, version: "1.3.0.dev20260710" }
+  kv-p2p-transfer: nixl
+  multinode: true
+  disagg: true
+  scenarios:
+    fixed-seq-len:
+    - isl: 8192
+      osl: 1024
+      search-space:
+      - spec-decoding: "mtp"
+        conc-list: [12, 24, 48]
+        prefill:
+          num-worker: 1
+          tp: 4
+          ep: 1
+          dp-attn: false
+          additional-settings:
+          - "CONFIG_FILE=recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p6d-tp4-tp4-eagle3-8k1k.yaml"
+        decode:
+          num-worker: 6
+          tp: 4
+          ep: 1
+          dp-attn: false
+      - spec-decoding: "mtp"
+        conc-list: [4, 8, 16, 32, 64]
+        prefill:
+          num-worker: 1
+          tp: 4
+          ep: 1
+          dp-attn: false
+          additional-settings:
+          - "CONFIG_FILE=recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p4d-tp4-tp4-eagle3-8k1k.yaml"
+        decode:
+          num-worker: 4
+          tp: 4
+          ep: 1
+          dp-attn: false
+      - spec-decoding: "mtp"
+        conc-list: [32, 64, 96]
+        prefill:
+          num-worker: 1
+          tp: 4
+          ep: 1
+          dp-attn: false
+          additional-settings:
+          - "CONFIG_FILE=recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p2d-tp4-tp4-eagle3-8k1k.yaml"
+        decode:
+          num-worker: 2
+          tp: 4
+          ep: 1
+          dp-attn: false
+      - spec-decoding: "mtp"
+        conc-list: [192, 256, 512]
+        prefill:
+          num-worker: 1
+          tp: 4
+          ep: 1
+          dp-attn: false
+          additional-settings:
+          - "CONFIG_FILE=recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p1d-tp4-tp4-eagle3-8k1k.yaml"
+        decode:
+          num-worker: 1
+          tp: 4
+          ep: 1
+          dp-attn: false
+
+      - spec-decoding: "mtp"
+        conc-list: [12, 24, 48]
+        prefill:
+          num-worker: 1
+          tp: 4
+          ep: 1
+          dp-attn: false
+          additional-settings:
+          - "CONFIG_FILE=recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p3d-tp4-tp4-eagle3-8k1k.yaml"
+        decode:
+          num-worker: 3
+          tp: 4
+          ep: 1
+          dp-attn: false
+
+
+      - spec-decoding: "mtp"
+        conc-list: [256, 512]
+        prefill:
+          num-worker: 1
+          tp: 4
+          ep: 1
+          dp-attn: false
+          additional-settings:
+          - "CONFIG_FILE=recipes/vllm/minimax-m3/b200-fp4/8k1k/mtp/1p1d-tp4-dep4-eagle3-8k1k.yaml"
+        decode:
+          num-worker: 1
+          tp: 4
+          ep: 4
+          dp-attn: true
+
 minimaxm3-fp4-b300-dynamo-vllm-mtp:
   image: vllm/vllm-openai:nightly-5e35a6f4f9bbc217c599692157ca985c894373f7
   model: nvidia/MiniMax-M3-NVFP4
diff --git a/perf-changelog.yaml b/perf-changelog.yaml
index b5aa7a385..43dc7fdb6 100644
--- a/perf-changelog.yaml
+++ b/perf-changelog.yaml
@@ -5425,3 +5425,9 @@
   description:
     - "Extend the search space to include the TP2EP1 configuration for fixed seq len 8k1k"
   pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/2446
+
+- config-keys:
+    - minimaxm3-fp4-b200-dynamo-vllm-mtp
+  description:
+    - "Add MiniMax M3 NVFP4 B200 disagg EAGLE3 MTP recipes, mirrored from B300"
+  pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/2462