[Doc] Fix typos, grammar, and broken commands across docs (#46398)

Signed-off-by: MichaelCaoo <[email protected]>
Co-authored-by: Claude <[email protected]>

docs/benchmarking/cli.md

@@ -338,7 +338,7 @@ vllm bench serve \
     --model meta-llama/Meta-Llama-3-8B-Instruct \
     --dataset-name spec_bench \
     --dataset-path "<YOUR_DOWNLOADED_PATH>/data/spec_bench/question.jsonl" \
-    --num-prompts -1
+    --num-prompts -1 \
     --spec-bench-category "summarization"
 ```
 
@@ -352,7 +352,7 @@ vllm bench serve \
 First, download the dataset to a folder, using this one liner:
 
 ```bash
-curl -LsSf https://raw.githubusercontent.com/NVIDIA-NeMo/Skills/refs/heads/main/nemo_skills/dataset/speed-bench/prepare.py \| python3 -
+curl -LsSf https://raw.githubusercontent.com/NVIDIA-NeMo/Skills/refs/heads/main/nemo_skills/dataset/speed-bench/prepare.py | python3 -
 ```
 
 The command supports also the following arguments:
@@ -388,7 +388,7 @@ vllm bench serve \
     --model meta-llama/Llama-3.3-70B-Instruct \
     --dataset-name speed_bench \
     --dataset-path "<YOUR_DOWNLOADED_PATH>/data/speed_bench" \
-    --num-prompts -1
+    --num-prompts -1 \
     --speed-bench-category "multilingual"
 ```
 
@@ -398,7 +398,7 @@ Run all categories in the Throughput split (2k ISL):
 vllm bench serve \
     --model meta-llama/Llama-3.3-70B-Instruct \
     --dataset-name speed_bench \
-    --speed-bench-dataset-subset throughput_2k
+    --speed-bench-dataset-subset throughput_2k \
     --dataset-path "<YOUR_DOWNLOADED_PATH>/data/speed_bench/" \
     --num-prompts -1
 ```

docs/configuration/optimization.md

@@ -109,7 +109,7 @@ from vllm import LLM
 
 # Combine pipeline and tensor parallelism
 llm = LLM(
-    model="meta-llama/Llama-3.3-70B-Instruct,
+    model="meta-llama/Llama-3.3-70B-Instruct",
     tensor_parallel_size=4,
     pipeline_parallel_size=2,
 )

docs/design/cuda_graphs.md

@@ -161,11 +161,11 @@ class AttentionCGSupport(enum.Enum):
     ALWAYS = 3
     """CUDA Graphs always supported; supports mixed-prefill-decode"""
     UNIFORM_BATCH = 2
-    """CUDA Graphs supported for batches the only contain query lengths that are
+    """CUDA Graphs supported for batches that only contain query lengths that are
     the same, this can be used for spec-decode 
         i.e. "decodes" are 1 + num_speculative_tokens"""
     UNIFORM_SINGLE_TOKEN_DECODE = 1
-    """CUDA Graphs supported for batches the only contain query_len==1 decodes"""
+    """CUDA Graphs supported for batches that only contain query_len==1 decodes"""
     NEVER = 0
     """NO CUDA Graphs support"""
 ```

docs/design/metrics.md

@@ -685,7 +685,7 @@ documentation for this option states:
 > use of possibly costly and or blocking operations and hence might
 > have a performance impact.
 
-The metrics were added by <https://github.com/vllm-project/vllm/pull/7089> and who up in an OpenTelemetry trace
+The metrics were added by <https://github.com/vllm-project/vllm/pull/7089> and show up in an OpenTelemetry trace
 as:
 
 ```text

docs/design/prefix_caching.md

@@ -27,7 +27,7 @@ In the example above, the KV cache in the first block can be uniquely identified
     For `vllm serve`, you can control the hashing algorithm via `--prefix-caching-hash-algo`:
     - `sha256` (default): Uses Python's `pickle` for serialization. Hashes may not be reproducible across different Python or vLLM versions.
     - `sha256_cbor`: Uses `cbor2` for serialization, providing a reproducible, cross-language compatible hash. This is recommended for deterministic caching across environments.
-    - `xxhash`: `Uses Pickle serialization with xxHash (128-bit) for faster, non-cryptographic hashing. Requires the optional `xxhash` package. IMPORTANT: Use of a hashing algorithm that is not considered cryptographically secure theoretically increases the risk of hash collisions, which can cause undefined behavior or even leak private information in multi-tenant environments. Even if collisions are still very unlikely, it is important to consider your security risk tolerance against the performance benefits before turning this on.
+    - `xxhash`: Uses Pickle serialization with xxHash (128-bit) for faster, non-cryptographic hashing. Requires the optional `xxhash` package. IMPORTANT: Use of a hashing algorithm that is not considered cryptographically secure theoretically increases the risk of hash collisions, which can cause undefined behavior or even leak private information in multi-tenant environments. Even if collisions are still very unlikely, it is important to consider your security risk tolerance against the performance benefits before turning this on.
     - `xxhash_cbor` combines canonical CBOR serialization with xxHash for reproducible hashing. Requires the optional `xxhash` package.    
 
 **A hashing example with multi-modality inputs**  
@@ -197,7 +197,7 @@ As can be seen, block 3 is a new full block and is cached. However, it is redund
 
 When a request is finished, we free all its blocks if no other requests are using them (reference count = 0). In this example, we free request 1 and block 2, 3, 4, 8 associated with it. We can see that the freed blocks are added to the tail of the free queue in the *reverse* order. This is because the last block of a request must hash more tokens and is less likely to be reused by other requests. As a result, it should be evicted first.
 
-![Free queue after a request us freed](../assets/design/prefix_caching/free.png)
+![Free queue after a request is freed](../assets/design/prefix_caching/free.png)
 
 ### Eviction (LRU)
 

docs/features/quantization/gptqmodel.md

@@ -55,7 +55,7 @@ Here is an example of how to quantize `meta-llama/Llama-3.2-1B-Instruct`:
 
 ## Running a quantized model with vLLM
 
-To run an GPTQModel quantized model with vLLM, you can use [DeepSeek-R1-Distill-Qwen-7B-gptqmodel-4bit-vortex-v2](https://huggingface.co/ModelCloud/DeepSeek-R1-Distill-Qwen-7B-gptqmodel-4bit-vortex-v2) with the following command:
+To run a GPTQModel quantized model with vLLM, you can use [DeepSeek-R1-Distill-Qwen-7B-gptqmodel-4bit-vortex-v2](https://huggingface.co/ModelCloud/DeepSeek-R1-Distill-Qwen-7B-gptqmodel-4bit-vortex-v2) with the following command:
 
 ```bash
 python examples/deployment/llm_engine_example.py \

docs/features/quantization/llm_compressor/int8_w8a8.md

@@ -78,8 +78,6 @@ def tokenize(sample):
 ds = ds.map(tokenize, remove_columns=ds.column_names)
 ```
 
-</details>
-
 ### 3. Applying Quantization
 
 Now, apply the quantization algorithms:

docs/features/speculative_decoding/README.md

@@ -190,7 +190,7 @@ For mitigation strategies, please refer to the FAQ entry *Can the output of a pr
 
 ## Known Feature Incompatibility
 
-1. Pipeline parallelism is not composible with speculative decoding as of `vllm<=0.15.0`
+1. Pipeline parallelism is not composable with speculative decoding as of `vllm<=0.15.0`
 2. Speculative decoding with a draft models is not supported in `vllm<=0.10.0`
 
 ## Resources for vLLM contributors

docs/features/tool_calling.md

@@ -338,7 +338,7 @@ Supported models:
 
 Flags: `--tool-call-parser deepseek_v31 --chat-template {see_above}`
 
-### OpenAI OSS Models ('openai`)
+### OpenAI OSS Models (`openai`)
 
 Supported models:
 

docs/models/pooling_models/README.md

@@ -143,7 +143,7 @@ enabling the corresponding APIs.
 
 The [classify][vllm.LLM.classify] method outputs a probability vector for each prompt.
 It is primarily designed for [classification models](classify.md).
-For more information about `LLM.embed`, see [this page](classify.md#offline-inference).
+For more information about `LLM.classify`, see [this page](classify.md#offline-inference).
 
 ### `LLM.embed`
 
@@ -302,7 +302,7 @@ Pooling models now support token-wise task.
 
 ### Score task
 
-`score` task have has been removed in v0.21, use `classify` instead. Only when a classification model outputs num_labels
+`score` task has been removed in v0.21, use `classify` instead. Only when a classification model outputs num_labels
 equal to 1 can it be used as a scoring model and have its scoring API enabled.
 
 ### Pooling multitask support

docs/models/pooling_models/scoring.md

@@ -440,7 +440,7 @@ More examples can be found here: [examples/pooling/score](../../../examples/pool
 
 ## Supported Features
 
-AS cross-encoder models are a subset of classification models that accept two prompts as input and output num_labels equal to 1, cross-encoder features should be consistent with (sequence) classification. For more information, see [this page](classify.md#supported-features).
+As cross-encoder models are a subset of classification models that accept two prompts as input and output num_labels equal to 1, cross-encoder features should be consistent with (sequence) classification. For more information, see [this page](classify.md#supported-features).
 
 ### Score Template