karthink · jdormit · Apr 7, 2025 · Jun 25, 2025 · Jun 25, 2025 · Jun 25, 2025
diff --git a/NEWS b/NEWS
@@ -7,8 +7,33 @@
 - Add support for ~gemini-2.5-pro~, ~gemini-2.5-flash~,
   ~gemini-2.5-flash-lite-preview-06-17~.
 
+- Add support for Open WebUI.  Open WebUI provides an
+  OpenAI-compatible API, so the "support" is just a new section of the
+  README with instructions.
+
 ** New features and UI changes
 
+- When including a file in the context, the abbreviated full path of
+  the file is included is now included instead of the basename.
+  Specifically, =/home/user/path/to/file= is included as
+  =~/path/to/file=.  This is to provide additional context for LLM
+  actions, including tool-use in subsequent conversation turns.  This
+  applies to context included via ~gptel-add~ or as a link in a
+  buffer.
+
+- Structured output support: ~gptel-request~ can now take an optional
+  schema argument to constrain LLM output to the specified JSON
+  schema.  The JSON schema can be provided as a serialized JSON string
+  or as an elisp object (a nested plist).  This feature works with all major
+  backends: OpenAI, Anthropic, Gemini, llama-cpp and Ollama.  It is
+  presently supported by some but not all "OpenAI-compatible API"
+  providers.  Note that this is only available via the ~gptel-request~
+  API, and currently unsupported by ~gptel-send~.
+
+- From the dry-run inspector buffer, you can now copy the Curl command
+  for the request.  Like when continuing the query, the request is
+  constructed from the contents of the buffer, which is editable.
+
 - gptel now handles Ollama models that return both reasoning content
   and tool calls in a single request.
 

diff --git a/README.org b/README.org
@@ -14,6 +14,7 @@ gptel is a simple Large Language Model chat client for Emacs, with support for m
 | Anthropic (Claude)   | ✓        | [[https://www.anthropic.com/api][API key]]                    |
 | Gemini               | ✓        | [[https://makersuite.google.com/app/apikey][API key]]                    |
 | Ollama               | ✓        | [[https://ollama.ai/][Ollama running locally]]     |
+| Open WebUI           | ✓        | [[https://openwebui.com/][Open WebUI running locally]]   |
 | Llama.cpp            | ✓        | [[https://github.com/ggml-org/llama.cpp/tree/master/tools/server#quick-start][Llama.cpp running locally]]  |
 | Llamafile            | ✓        | [[https://github.com/Mozilla-Ocho/llamafile#quickstart][Local Llamafile server]]     |
 | GPT4All              | ✓        | [[https://gpt4all.io/index.html][GPT4All running locally]]    |
@@ -24,6 +25,7 @@ gptel is a simple Large Language Model chat client for Emacs, with support for m
 | Mistral Le Chat      | ✓        | [[https://console.mistral.ai/api-keys][API key]]                    |
 | Perplexity           | ✓        | [[https://docs.perplexity.ai/docs/getting-started][API key]]                    |
 | OpenRouter           | ✓        | [[https://openrouter.ai/keys][API key]]                    |
+| AI/ML API            | ✓        | [[https://aimlapi.com/app/?utm_source=gptel&utm_medium=github&utm_campaign=integration][API key]]                    |
 | together.ai          | ✓        | [[https://api.together.xyz/settings/api-keys][API key]]                    |
 | Anyscale             | ✓        | [[https://docs.endpoints.anyscale.com/][API key]]                    |
 | PrivateGPT           | ✓        | [[https://github.com/zylon-ai/private-gpt#-documentation][PrivateGPT running locally]] |
@@ -101,6 +103,7 @@ gptel uses Curl if available, but falls back to the built-in url-retrieve to wor
       - [[#azure][Azure]]
       - [[#gpt4all][GPT4All]]
       - [[#ollama][Ollama]]
+      - [[#open-webui][Open WebUI]]
       - [[#gemini][Gemini]]
       - [[#llamacpp-or-llamafile][Llama.cpp or Llamafile]]
       - [[#kagi-fastgpt--summarizer][Kagi (FastGPT & Summarizer)]]
@@ -118,6 +121,7 @@ gptel uses Curl if available, but falls back to the built-in url-retrieve to wor
       - [[#github-models][Github Models]]
       - [[#novita-ai][Novita AI]]
       - [[#xai][xAI]]
+      - [[#aiml-api][AI/ML API]]
       - [[#github-copilotchat][Github CopilotChat]]
       - [[#aws-bedrock][AWS Bedrock]]
   - [[#usage][Usage]]
@@ -333,6 +337,62 @@ The above code makes the backend available to select.  If you want it to be the
 
 #+html: </details>
 
+#+html: <details><summary>
+**** Open WebUI
+#+html: </summary>
+
+[[https://openwebui.com/][Open WebUI]] is an open source, self-hosted system which provides a multi-user web chat interface and an API endpoint for accessing LLMs, especially LLMs running locally on inference servers like Ollama.
+
+Because it presents an OpenAI-compatible endpoint, you use ~gptel-make-openai~ to register it as a backend.
+
+For instance, you can use this form to register a backend for a local instance of Open Web UI served via http on port 3000:
+
+#+begin_src emacs-lisp
+(gptel-make-openai "OpenWebUI"
+  :host "localhost:3000"
+  :protocol "http"
+  :key "KEY_FOR_ACCESSING_OPENWEBUI"
+  :endpoint "/api/chat/completions"
+  :stream t
+  :models '("gemma3n:latest"))
+#+end_src
+
+Or if you are running Open Web UI on another host on your local network (~box.local~), serving via https with self-signed certificates, this will work:
+
+#+begin_src emacs-lisp
+(gptel-make-openai "OpenWebUI"
+  :host "box.local"
+  :curl-args '("--insecure") ; needed for self-signed certs
+  :key "KEY_FOR_ACCESSING_OPENWEBUI"
+  :endpoint "/api/chat/completions"
+  :stream t
+  :models '("gemma3n:latest"))
+#+end_src
+
+To find your API key in Open WebUI, click the user name in the bottom left, Settings, Account, and then Show by API Keys section.
+
+Refer to the documentation of =gptel-make-openai= for more configuration options.
+
+You can pick this backend from the menu when using gptel (see [[#usage][Usage]])
+
+***** (Optional) Set as the default gptel backend
+
+The above code makes the backend available to select.  If you want it to be the default backend for gptel, you can set this as the value of =gptel-backend=.  Use this instead of the above.
+#+begin_src emacs-lisp
+;; OPTIONAL configuration
+(setq
+ gptel-model "gemma3n:latest"
+ gptel-backend (gptel-make-openai "OpenWebUI"
+                 :host "localhost:3000"
+                 :protocol "http"
+                 :key "KEY_FOR_ACCESSING_OPENWEBUI"
+                 :endpoint "/api/chat/completions"
+                 :stream t
+                 :models '("gemma3n:latest")))
+#+end_src
+
+#+html: </details>
+
 #+html: <details><summary>
 **** Gemini
 #+html: </summary>
@@ -924,6 +984,41 @@ The above code makes the backend available to select.  If you want it to be the
 
 #+html: </details>
 
+#+html: <details><summary>
+**** AI/ML API
+#+html: </summary>
+
+AI/ML API provides 300+ AI models including Deepseek, Gemini, ChatGPT. The models run at enterprise-grade rate limits and uptimes.
+
+Register a backend with
+#+begin_src emacs-lisp
+;; AI/ML API offers an OpenAI compatible API
+(gptel-make-openai "AI/ML API"        ;Any name you want
+  :host "api.aimlapi.com"
+  :endpoint "/v1/chat/completions"
+  :stream t
+  :key "your-api-key"                ;can be a function that returns the key
+  :models '(deepseek-chat gemini-pro gpt-4o))
+#+end_src
+
+You can pick this backend from the menu when using gptel (see [[#usage][Usage]]).
+
+***** (Optional) Set as the default gptel backend
+
+The above code makes the backend available to select.  If you want it to be the default backend for gptel, you can set this as the value of =gptel-backend=.  Use this instead of the above.
+#+begin_src emacs-lisp
+;; OPTIONAL configuration
+(setq gptel-model 'gpt-4o
+      gptel-backend
+      (gptel-make-openai "AI/ML API"
+        :host "api.aimlapi.com"
+        :endpoint "/v1/chat/completions"
+        :stream t
+        :key "your-api-key"
+        :models '(deepseek-chat gemini-pro gpt-4o)))
+#+end_src
+
+#+html: </details>
 #+html: <details><summary>
 **** Github CopilotChat
 #+html: </summary>
@@ -1526,13 +1621,15 @@ Other Emacs clients for LLMs prescribe the format of the interaction (a comint s
   all)
 #+end_src
 
-|----------------------+--------------------------------------------------------------------|
-| *Connection options* |                                                                    |
-|----------------------+--------------------------------------------------------------------|
-| =gptel-use-curl=     | Use Curl (default), fallback to Emacs' built-in =url=.             |
-| =gptel-proxy=        | Proxy server for requests, passed to curl via =--proxy=.           |
-| =gptel-api-key=      | Variable/function that returns the API key for the active backend. |
-|----------------------+--------------------------------------------------------------------|
+|-------------------------+--------------------------------------------------------------------|
+| *Connection options*    |                                                                    |
+|-------------------------+--------------------------------------------------------------------|
+| =gptel-use-curl=        | Use Curl? (default), fallback to Emacs' built-in =url=.            |
+|                         | You can also specify the Curl path here.                           |
+| =gptel-proxy=           | Proxy server for requests, passed to curl via =--proxy=.           |
+| =gptel-curl-extra-args= | Extra arguments passed to Curl.                                    |
+| =gptel-api-key=         | Variable/function that returns the API key for the active backend. |
+|-------------------------+--------------------------------------------------------------------|
 
 |-----------------------+---------------------------------------------------------|
 | *LLM request options* | /(Note: not supported uniformly across LLMs)/           |
@@ -1677,10 +1774,17 @@ There are several more: [[https://github.com/iwahbe/chat.el][chat.el]], [[https:
 gptel is a general-purpose package for chat and ad-hoc LLM interaction.  The following packages use gptel to provide additional or specialized functionality:
 
 - [[https://github.com/karthink/gptel-quick][gptel-quick]]: Quickly look up the region or text at point.
+- [[https://github.com/jwiegley/gptel-prompts][gptel-prompts]]: System prompt manager for gptel.
+- [[https://github.com/dolmens/gptel-aibo/][gptel-aibo]]: A writing assistant system built on top of gptel.
+- [[https://github.com/kmontag/macher][Macher]]: Project-aware multi-file LLM editing for Emacs, based on gptel.
 - [[https://github.com/daedsidog/evedel][Evedel]]: Instructed LLM Programmer/Assistant.
 - [[https://github.com/lanceberge/elysium][Elysium]]: Automatically apply AI-generated changes as you code.
+- [[https://github.com/jwiegley/ob-gptel][ob-gptel]]: Org-babel backend for running gptel queries.
+- [[https://github.com/JDNdeveloper/gptel-autocomplete][gptel-autocomplete]]: Inline completions using gptel.
 - [[https://github.com/kamushadenes/ai-blog.el][ai-blog.el]]: Streamline generation of blog posts in Hugo.
-- [[https://github.com/douo/magit-gptcommit][magit-gptcommit]]: Generate Commit Messages within magit-status Buffer using gptel.
+- [[https://github.com/lakkiy/gptel-commit][gptel-commit]]: Generate commit messages using gptel.
+- [[https://github.com/douo/magit-gptcommit][magit-gptcommit]]: Generate commit messages within magit-status Buffer using gptel.
+- [[https://github.com/ragnard/gptel-magit/][gptel-magit]]: Generate commit messages for magit using gptel.
 - [[https://github.com/armindarvish/consult-omni][consult-omni]]: Versatile multi-source search package.  It includes gptel as one of its many sources.
 - [[https://github.com/ultronozm/ai-org-chat.el][ai-org-chat]]: Provides branching conversations in Org buffers using gptel.  (Note that gptel includes this feature as well (see =gptel-org-branching-context=), but requires a recent version of Org mode 9.7 or later to be installed.)
 - [[https://github.com/rob137/Corsair][Corsair]]: Helps gather text to populate LLM prompts for gptel.

diff --git a/gptel-anthropic.el b/gptel-anthropic.el
@@ -230,13 +230,29 @@ Mutate state INFO with response metadata."
                      (gptel--model-capable-p 'cache))
             (nconc (aref tools-array (1- (length tools-array)))
                    '(:cache_control (:type "ephemeral")))))))
+    (when gptel--schema
+      (plist-put prompts-plist :tools
+                 (vconcat
+                  (list (gptel--parse-schema backend gptel--schema))
+                  (plist-get prompts-plist :tools)))
+      (plist-put prompts-plist :tool_choice
+                 `(:type "tool" :name ,gptel--ersatz-json-tool)))
     ;; Merge request params with model and backend params.
     (gptel--merge-plists
      prompts-plist
      gptel--request-params
      (gptel-backend-request-params gptel-backend)
      (gptel--model-request-params  gptel-model))))
 
+(cl-defmethod gptel--parse-schema ((_backend gptel-anthropic) schema)
+  ;; Unlike the other backends, Anthropic generates JSON using a tool call.  We
+  ;; write the tool here, meant to be added to :tools.
+  (list
+   :name "response_json"
+   :description "Record JSON output according to user prompt"
+   :input_schema (gptel--preprocess-schema
+                  (gptel--dispatch-schema-type schema))))
+
 (cl-defmethod gptel--parse-tools ((_backend gptel-anthropic) tools)
   "Parse TOOLS to the Anthropic API tool definition spec.
 

diff --git a/gptel-curl.el b/gptel-curl.el
@@ -112,8 +112,6 @@ REQUEST-DATA is the data to send, TOKEN is a unique identifier."
               collect (format "-H%s: %s" key val))
      (list url))))
 
-;;TODO: The :transformer argument here is an alternate implementation of
-;;`gptel-response-filter-functions'. The two need to be unified.
 ;;;###autoload
 (defun gptel-curl-get-response (fsm)
   "Fetch response to prompt in state FSM from the LLM using Curl.
@@ -283,6 +281,9 @@ Optional RAW disables text properties and transformation."
              (set-marker-insertion-type tracking-marker t)
              (plist-put info :tracking-marker tracking-marker))
            (goto-char tracking-marker)
+           (when (plist-get info :last-was-tool-result)
+             (insert gptel-response-separator)
+             (plist-put info :last-was-tool-result nil))
            (unless raw
              (when transformer
                (setq response (funcall transformer response)))
@@ -297,7 +298,8 @@ Optional RAW disables text properties and transformation."
     (`(tool-call . ,tool-calls)
      (gptel--display-tool-calls tool-calls info))
     (`(tool-result . ,tool-results)
-     (gptel--display-tool-results tool-results info))))
+     (gptel--display-tool-results tool-results info)
+     (plist-put info :last-was-tool-result t))))
 
 (defun gptel-curl--stream-filter (process output)
   (let* ((fsm (car (alist-get process gptel--request-alist)))
@@ -366,18 +368,17 @@ Optional RAW disables text properties and transformation."
                       (progn (setq response (cons 'reasoning response))
                              (plist-put proc-info :reasoning-block 'in))
                     (plist-put proc-info :reasoning-block 'done)))
-                 ((length> response 0)
+                 ((and (not (eq reasoning-block t)) (length> response 0))
                   (if-let* ((idx (string-match-p "</think>" response)))
                       (progn
                         (funcall callback
                                  (cons 'reasoning ;last reasoning chunk
                                        (string-trim-left
                                         (substring response nil (+ idx 8))))
                                  proc-info)
-                        ;; Signal end of reasoning stream
-                        (funcall callback '(reasoning . t) proc-info)
-                        (setq response (substring response (+ idx 8)))
-                        (plist-put proc-info :reasoning-block 'done))
+                        (setq reasoning-block t) ;Signal end of reasoning stream
+                        (plist-put proc-info :reasoning-block t)
+                        (setq response (substring response (+ idx 8))))
                     (setq response (cons 'reasoning response)))))
                 (when (eq reasoning-block t) ;End of reasoning block
                   (funcall callback '(reasoning . t) proc-info)