docs: document tool calling on vLLM and MLX backends

openai-functions.md used to claim LocalAI tool calling worked only on
llama.cpp-compatible models. That was true when it was written; it's
not true now — vLLM (since PR #9328) and MLX/MLX-VLM both extract
structured tool calls from model output.

- openai-functions.md: new 'Supported backends' matrix covering
  llama.cpp, vllm, vllm-omni, mlx, mlx-vlm, with the key distinction
  that vllm needs an explicit tool_parser: option while mlx auto-
  detects from the chat template. Reasoning content (think tags) is
  extracted on the same set of backends. Added setup snippets for
  both the vllm and mlx paths, and noted the gallery importer
  pre-fills tool_parser:/reasoning_parser: for known families.
- compatibility-table.md: fix the stale 'Streaming: no' for vllm,
  vllm-omni, mlx, mlx-vlm (all four support streaming now). Add
  'Functions' to their capabilities. Also widen the MLX Acceleration
  column to reflect the CPU/CUDA/Jetson L4T backends that already
  exist in backend/index.yaml — 'Metal' on its own was misleading.

Author: mudler

docs/content/features/openai-functions.md

@@ -6,21 +6,70 @@ weight = 17
 url = "/features/openai-functions/"
 +++
 
-LocalAI supports running OpenAI [functions and tools API](https://platform.openai.com/docs/api-reference/chat/create#chat-create-tools) with `llama.cpp` compatible models.
+LocalAI supports running the OpenAI [functions and tools API](https://platform.openai.com/docs/api-reference/chat/create#chat-create-tools) across multiple backends. The OpenAI request shape is the same regardless of which backend runs your model — LocalAI is responsible for extracting structured tool calls from the model's output before returning the response.
 
 ![localai-functions-1](https://github.com/ggerganov/llama.cpp/assets/2420543/5bd15da2-78c1-4625-be90-1e938e6823f1)
 
 To learn more about OpenAI functions, see also the [OpenAI API blog post](https://openai.com/blog/function-calling-and-other-api-updates).
 
-LocalAI is also supporting [JSON mode](https://platform.openai.com/docs/guides/text-generation/json-mode) out of the box with llama.cpp-compatible models.
+LocalAI also supports [JSON mode](https://platform.openai.com/docs/guides/text-generation/json-mode) out of the box on llama.cpp-compatible models.
 
-💡 Check out also [LocalAGI](https://github.com/mudler/LocalAGI) for an example on how to use LocalAI functions.
+💡 Check out [LocalAGI](https://github.com/mudler/LocalAGI) for an example on how to use LocalAI functions.
+
+## Supported backends
+
+| Backend | How tool calls are extracted |
+|---------|------------------------------|
+| `llama.cpp` | C++ incremental parser; any `ggml`/`gguf` model works out of the box, no configuration needed |
+| `vllm` | vLLM's native `ToolParserManager` — select a parser with `tool_parser:<name>` in the model `options`. Auto-set by the gallery importer for known families |
+| `vllm-omni` | Same as vLLM |
+| `mlx` | `mlx_lm.tool_parsers` — **auto-detected from the chat template**, no configuration needed |
+| `mlx-vlm` | `mlx_vlm.tool_parsers` (with fallback to mlx-lm parsers) — **auto-detected from the chat template**, no configuration needed |
+
+Reasoning content (`<think>...</think>` blocks from DeepSeek R1, Qwen3, Gemma 4, etc.) is returned in the OpenAI `reasoning_content` field on the same backends.
 
 ## Setup
 
-OpenAI functions are available only with `ggml` or `gguf` models compatible with `llama.cpp`.
+### llama.cpp
+
+No configuration required — the autoparser detects the tool call format for any `ggml`/`gguf` model that was trained with tool support.
+
+### vLLM / vLLM Omni
+
+The parser must be specified explicitly because vLLM itself doesn't auto-detect one. Pass it via the model `options`:
+
+```yaml
+name: qwen3-8b
+backend: vllm
+parameters:
+  model: Qwen/Qwen3-8B
+options:
+  - tool_parser:hermes
+  - reasoning_parser:qwen3
+template:
+  use_tokenizer_template: true
+```
+
+When you import a vLLM model through the LocalAI gallery, the importer looks up the model family and pre-fills `tool_parser:` and `reasoning_parser:` for you — you only need to override them for non-standard model names.
+
+Available tool parsers include `hermes`, `llama3_json`, `llama4_pythonic`, `mistral`, `qwen3_xml`, `deepseek_v3`, `granite4`, `kimi_k2`, `glm45`, and more. Available reasoning parsers include `deepseek_r1`, `qwen3`, `mistral`, `gemma4`, `granite`. See the upstream vLLM documentation for the full list.
+
+### MLX / MLX-VLM
+
+MLX backends **auto-detect** the right tool parser by inspecting the model's chat template — you don't need to set anything. Just load an MLX-quantized model that was trained with tool support:
+
+```yaml
+name: qwen2.5-0.5b-mlx
+backend: mlx
+parameters:
+  model: mlx-community/Qwen2.5-0.5B-Instruct-4bit
+template:
+  use_tokenizer_template: true
+```
+
+The gallery importer will still append `tool_parser:` and `reasoning_parser:` entries to the YAML for visibility and consistency with the other backends, but those are informational — the runtime auto-detection in the MLX backend ignores them and uses the parser matched to the chat template.
 
-You don't need to do anything specific - just use `ggml` or `gguf` models.
+Supported parser families: `hermes`/`json_tools`, `mistral`, `gemma4`, `glm47`, `kimi_k2`, `longcat`, `minimax_m2`, `pythonic`, `qwen3_coder`, `function_gemma`.
 
 
 ## Usage example

docs/content/reference/compatibility-table.md

@@ -20,11 +20,11 @@ LocalAI will attempt to automatically load models which are not explicitly confi
 |---------|-------------|------------|------------|-----------|-------------|
 | [llama.cpp](https://github.com/ggerganov/llama.cpp) | LLM inference in C/C++. Supports LLaMA, Mamba, RWKV, Falcon, Starcoder, GPT-2, [and many others](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#description) | GPT, Functions | yes | yes | CPU, CUDA 12/13, ROCm, Intel SYCL, Vulkan, Metal, Jetson L4T |
 | [ik_llama.cpp](https://github.com/ikawrakow/ik_llama.cpp) | Hard fork of llama.cpp optimized for CPU/hybrid CPU+GPU with IQK quants, custom quant mixes, and MLA for DeepSeek | GPT | yes | yes | CPU (AVX2+) |
-| [vLLM](https://github.com/vllm-project/vllm) | Fast LLM serving with PagedAttention | GPT | no | no | CUDA 12, ROCm, Intel |
-| [vLLM Omni](https://github.com/vllm-project/vllm) | Unified multimodal generation (text, image, video, audio) | Multimodal GPT | no | no | CUDA 12, ROCm |
+| [vLLM](https://github.com/vllm-project/vllm) | Fast LLM serving with PagedAttention | GPT, Functions | no | yes | CPU, CUDA 12, ROCm, Intel |
+| [vLLM Omni](https://github.com/vllm-project/vllm) | Unified multimodal generation (text, image, video, audio) | Multimodal GPT, Functions | no | yes | CUDA 12, ROCm |
 | [transformers](https://github.com/huggingface/transformers) | HuggingFace Transformers framework | GPT, Embeddings, Multimodal | yes | yes* | CPU, CUDA 12/13, ROCm, Intel, Metal |
-| [MLX](https://github.com/ml-explore/mlx-lm) | Apple Silicon LLM inference | GPT | no | no | Metal |
-| [MLX-VLM](https://github.com/Blaizzy/mlx-vlm) | Vision-Language Models on Apple Silicon | Multimodal GPT | no | no | Metal |
+| [MLX](https://github.com/ml-explore/mlx-lm) | Apple Silicon LLM inference | GPT, Functions | no | yes | Metal, CPU, CUDA 12/13, Jetson L4T |
+| [MLX-VLM](https://github.com/Blaizzy/mlx-vlm) | Vision-Language Models on Apple Silicon | Multimodal GPT, Functions | no | yes | Metal, CPU, CUDA 12/13, Jetson L4T |
 | [MLX Distributed](https://github.com/ml-explore/mlx-lm) | Distributed LLM inference across multiple Apple Silicon Macs | GPT | no | no | Metal |
 
 ## Speech-to-Text