diff --git a/gptel-anthropic.el b/gptel-anthropic.el
index 9dd2214f..975c6262 100644
--- a/gptel-anthropic.el
+++ b/gptel-anthropic.el
@@ -468,45 +468,58 @@ files in the context."
 ;;                   (t context-string)))
 ;;         (plist-get (car (last prompts)) :content)))
 
+(defconst gptel--anthropic-cost-url "https://docs.anthropic.com/en/docs/about-claude/models#model-comparison-table"
+  "Web page for current Anthropic model pricing.")
+
 (defconst gptel--anthropic-models
-  '((claude-3-7-sonnet-20250219
+  `((claude-3-7-sonnet-20250219
      :description "Hybrid model capable of standard thinking and extended thinking modes"
      :capabilities (media tool-use cache)
      :mime-types ("image/jpeg" "image/png" "image/gif" "image/webp" "application/pdf")
      :context-window 200
+     :max-output 8192
      :input-cost 3
      :output-cost 15
+     :cost-url ,gptel--anthropic-cost-url
      :cutoff-date "2025-02")
     (claude-3-5-sonnet-20241022
      :description "Highest level of intelligence and capability"
      :capabilities (media tool-use cache)
      :mime-types ("image/jpeg" "image/png" "image/gif" "image/webp" "application/pdf")
      :context-window 200
+     :max-output 8192
      :input-cost 3
      :output-cost 15
+     :cost-url ,gptel--anthropic-cost-url
      :cutoff-date "2024-04")
     (claude-3-5-sonnet-20240620
      :description "Highest level of intelligence and capability (earlier version)"
      :capabilities (media tool-use cache)
      :mime-types ("image/jpeg" "image/png" "image/gif" "image/webp")
      :context-window 200
+     :max-output 8192
      :input-cost 3
      :output-cost 15
+     :cost-url ,gptel--anthropic-cost-url
      :cutoff-date "2024-04")
     (claude-3-5-haiku-20241022
      :description "Intelligence at blazing speeds"
      :capabilities (tool-use cache)
      :context-window 200
+     :max-output 8192
      :input-cost 1.00
      :output-cost 5.00
+     :cost-url ,gptel--anthropic-cost-url
      :cutoff-date "2024-07")
     (claude-3-opus-20240229
      :description "Top-level performance, intelligence, fluency, and understanding"
      :capabilities (media tool-use cache)
      :mime-types ("image/jpeg" "image/png" "image/gif" "image/webp")
      :context-window 200
+     :max-output 4096
      :input-cost 15
      :output-cost 75
+     :cost-url ,gptel--anthropic-cost-url
      :cutoff-date "2023-08")
     (claude-3-sonnet-20240229
      :description "Balance of intelligence and speed (legacy model)"
@@ -515,13 +528,16 @@ files in the context."
      :context-window 200
      :input-cost 3
      :output-cost 15
+     :cost-url ,gptel--anthropic-cost-url
      :cutoff-date "2023-08")
     (claude-3-haiku-20240307
      :description "Fast and most compact model for near-instant responsiveness"
      :capabilities (tool-use cache)
      :context-window 200
+     :max-output 4096
      :input-cost 0.25
      :output-cost 1.25
+     :cost-url ,gptel--anthropic-cost-url
      :cutoff-date "2023-08"))
   "List of available Anthropic models and associated properties.
 Keys:
@@ -534,10 +550,14 @@ Keys:
 
 - `:context-window': the context window size, in thousands of tokens.
 
+- `:max-output': the maximum number of output tokens.
+
 - `:input-cost': the input cost, in US dollars per million tokens.
 
 - `:output-cost': the output cost, in US dollars per million tokens.
 
+- `:cost-url': web page for current model pricing.
+
 - `:cutoff-date': the knowledge cutoff date.
 
 - `:request-params': a plist of additional request parameters to
diff --git a/gptel-gemini.el b/gptel-gemini.el
index aa75c82f..cb291513 100644
--- a/gptel-gemini.el
+++ b/gptel-gemini.el
@@ -313,16 +313,21 @@ files in the context."
                   current))
         (plist-get (car (last prompts)) :parts))))
 
+(defconst gptel--gemini-cost-url "https://ai.google.dev/pricing"
+  "Web page for current Gemini model pricing.")
+
 (defconst gptel--gemini-models
-  '((gemini-1.5-pro-latest
+  `((gemini-1.5-pro-latest
      :description "Google's latest model with enhanced capabilities across various tasks"
      :capabilities (tool-use json media)
      :mime-types ("image/png" "image/jpeg" "image/webp" "image/heic" "image/heif"
                   "application/pdf" "text/plain" "text/csv" "text/html")
      :context-window 2000
+     :max-output 8192
      ;; input & output price is halved for prompts of 128k tokens or less
      :input-cost 2.50
      :output-cost 10
+     :cost-url ,gptel--gemini-cost-url
      :cutoff-date "2024-05")
     (gemini-2.0-flash-exp
      :description "Next generation features, superior speed, native tool use"
@@ -337,19 +342,23 @@ files in the context."
      :mime-types ("image/png" "image/jpeg" "image/webp" "image/heic" "image/heif"
                   "application/pdf" "text/plain" "text/csv" "text/html")
      :context-window 1000
+     :max-output 8192
      ;; input & output price is halved for prompts of 128k tokens or less
      :input-cost 0.15
      :output-cost 0.60
+     :cost-url ,gptel--gemini-cost-url
      :cutoff-date "2024-05")
     (gemini-1.5-flash-8b
      :description "High volume and lower intelligence tasks"
      :capabilities (tool-use json media)
      :context-window 1000
+     :max-output 8192
      :mime-types ("image/png" "image/jpeg" "image/webp" "image/heic" "image/heif"
                   "application/pdf" "text/plain" "text/csv" "text/html")
      ;; input & output price is halved for prompts of 128k tokens or less
      :input-cost 0.075
      :output-cost 0.30
+     :cost-url ,gptel--gemini-cost-url
      :cutoff-date "2024-10")
     (gemini-exp-1206
      :description "Improved coding, reasoning and vision capabilities"
@@ -363,8 +372,10 @@ files in the context."
      :mime-types ("image/png" "image/jpeg" "image/webp" "image/heic" "image/heif"
                   "application/pdf" "text/plain" "text/csv" "text/html")
      :context-window 1000
+     :max-output 8192
      :input-cost 0.10
      :output-cost 0.40
+     :cost-url ,gptel--gemini-cost-url
      :cutoff-date "2024-08")
     (gemini-2.0-flash-lite-preview-02-05
      :description "Gemini 2.0 Flash model optimized for cost efficiency and low latency"
@@ -372,6 +383,7 @@ files in the context."
      :context-window 1000
      :input-cost 0.075
      :output-cost 0.30
+     :cost-url ,gptel--gemini-cost-url
      :cutoff-date "2024-08")
     (gemini-2.0-pro-exp-02-05
      :description "Next gen, high speed, multimodal for a diverse variety of tasks"
@@ -379,12 +391,14 @@ files in the context."
      :context-window 2000
      :input-cost 0.00
      :output-cost 0.00
+     :cost-url ,gptel--gemini-cost-url
      :cutoff-date "2024-08")
     (gemini-2.0-flash-thinking-exp-01-21
      :description "Next gen, high speed, multimodal for a diverse variety of tasks"
      :capabilities (json)
      :input-cost 0.00
      :output-cost 0.00
+     :cost-url ,gptel--gemini-cost-url
      :cutoff-date "2024-08")
     (gemini-2.0-flash-exp
      :description "Multi-modal, streaming, tool use 2000 RPM"
@@ -394,6 +408,7 @@ files in the context."
      :context-window 1000
      :input-cost 0.00
      :output-cost 0.00
+     :cost-url ,gptel--gemini-cost-url
      :cutoff-date "2024-08")
     (gemini-2.0-flash-thinking-exp
      :description "DEPRECATED: Please use gemini-2.0-flash-thinking-exp-01-21 instead."
@@ -413,10 +428,14 @@ Keys:
 
 - `:context-window': the context window size, in thousands of tokens.
 
+- `:max-output': maximum number of output tokens.
+
 - `:input-cost': the input cost, in US dollars per million tokens.
 
 - `:output-cost': the output cost, in US dollars per million tokens.
 
+- `:cost-url': web page for current model pricing.
+
 - `:cutoff-date': the knowledge cutoff date.
 
 - `:request-params': a plist of additional request parameters to
diff --git a/gptel-openai-extras.el b/gptel-openai-extras.el
index c9e06307..e17a1f30 100644
--- a/gptel-openai-extras.el
+++ b/gptel-openai-extras.el
@@ -339,12 +339,16 @@ The Deepseek API requires strictly alternating roles (user/assistant) in message
           (models '((deepseek-reasoner
                      :capabilities (tool reasoning)
                      :context-window 64
+		     :max-output 8192
                      :input-cost 0.55
-                     :output-cost 2.19)
+                     :output-cost 2.19
+		     :cost-url "https://api-docs.deepseek.com/quick_start/pricing/")
                     (deepseek-chat
                      :capabilities (tool)
                      :context-window 64
+		     :max-output 8192
                      :input-cost 0.27
+		     :cost-url "https://api-docs.deepseek.com/quick_start/pricing/"
                      :output-cost 1.10))))
   "Register a DeepSeek backend for gptel with NAME."
   (declare (indent 1))
diff --git a/gptel-transient.el b/gptel-transient.el
index 24dce163..e316a6d4 100644
--- a/gptel-transient.el
+++ b/gptel-transient.el
@@ -576,6 +576,13 @@ Also format its value in the Transient menu."
     (gptel--infix-context-add-file)
     (gptel--infix-context-remove-all)
     (gptel--suffix-context-buffer)]
+   ["" "Info"
+    ("h" "Describe model"
+     (lambda ()
+       "Describe the current gptel model."
+       (interactive)
+       (describe-gptel-model gptel-backend gptel-model))
+     :transient t)]
    [:pad-keys t
     :if (lambda () (and gptel-use-tools gptel--known-tools))
     "" (:info
diff --git a/gptel.el b/gptel.el
index e8f5e140..171aca02 100644
--- a/gptel.el
+++ b/gptel.el
@@ -520,14 +520,18 @@ the same as t."
 
 (defvar gptel--known-backends)
 
+(defconst gptel--openai-cost-url "https://platform.openai.com/docs/pricing"
+  "Web page for current OpenAI model pricing.")
+
 (defconst gptel--openai-models
-  '((gpt-4o
+  `((gpt-4o
      :description "Advanced model for complex tasks; cheaper & faster than GPT-Turbo"
      :capabilities (media tool-use json url)
      :mime-types ("image/jpeg" "image/png" "image/gif" "image/webp")
      :context-window 128
      :input-cost 2.50
      :output-cost 10
+     :cost-url ,gptel--openai-cost-url
      :cutoff-date "2023-10")
     (gpt-4o-mini
      :description "Cheap model for fast tasks; cheaper & more capable than GPT-3.5 Turbo"
@@ -536,6 +540,7 @@ the same as t."
      :context-window 128
      :input-cost 0.15
      :output-cost 0.60
+     :cost-url ,gptel--openai-cost-url
      :cutoff-date "2023-10")
     (gpt-4-turbo
      :description "Previous high-intelligence model"
@@ -544,6 +549,7 @@ the same as t."
      :context-window 128
      :input-cost 10
      :output-cost 30
+     :cost-url ,gptel--openai-cost-url
      :cutoff-date "2023-12")
     ;; points to gpt-4-0613
     (gpt-4
@@ -553,6 +559,7 @@ the same as t."
      :context-window 8.192
      :input-cost 30
      :output-cost 60
+     :cost-url ,gptel--openai-cost-url
      :cutoff-date "2023-09")
     (gpt-4-turbo-preview
      :description "Points to gpt-4-0125-preview"
@@ -561,6 +568,7 @@ the same as t."
      :context-window 128
      :input-cost 10
      :output-cost 30
+     :cost-url ,gptel--openai-cost-url
      :cutoff-date "2023-12")
     (gpt-4-0125-preview
      :description "GPT-4 Turbo preview model intended to reduce cases of “laziness”"
@@ -569,6 +577,7 @@ the same as t."
      :context-window 128
      :input-cost 10
      :output-cost 30
+     :cost-url ,gptel--openai-cost-url
      :cutoff-date "2023-12")
     (gpt-4.5-preview
      :description "Largest and most capable GPT model to date"
@@ -577,6 +586,7 @@ the same as t."
      :context-window 128
      :input-cost 75
      :output-cost 150
+     :cost-url ,gptel--openai-cost-url
      :cutoff-date "2023-10")
     (o1
      :description "Reasoning model designed to solve hard problems across domains"
@@ -586,6 +596,7 @@ the same as t."
      :input-cost 15
      :output-cost 60
      :cutoff-date "2023-10"
+     :cost-url ,gptel--openai-cost-url
      :request-params (:stream :json-false))
     (o1-preview
      :description "DEPRECATED: PLEASE USE o1"
@@ -594,6 +605,7 @@ the same as t."
      :context-window 128
      :input-cost 15
      :output-cost 60
+     :cost-url ,gptel--openai-cost-url
      :cutoff-date "2023-10"
      :capabilities (nosystem reasoning)
      :request-params (:stream :json-false))
@@ -602,6 +614,7 @@ the same as t."
      :context-window 128
      :input-cost 3
      :output-cost 12
+     :cost-url ,gptel--openai-cost-url
      :cutoff-date "2023-10"
      :capabilities (nosystem reasoning)
      :request-params (:stream :json-false))
@@ -610,6 +623,7 @@ the same as t."
      :context-window 200
      :input-cost 3
      :output-cost 12
+     :cost-url ,gptel--openai-cost-url
      :cutoff-date "2023-10"
      :capabilities (nosystem reasoning)
      :request-params (:stream :json-false))
@@ -617,13 +631,15 @@ the same as t."
     (gpt-4-32k
      :capabilities (tool-use)
      :input-cost 60
-     :output-cost 120)
+     :output-cost 120
+     :cost-url ,gptel--openai-cost-url)
     (gpt-4-1106-preview
      :description "Preview model with improved function calling support"
      :capabilities (tool-use)
      :context-window 128
      :input-cost 10
      :output-cost 30
+     :cost-url ,gptel--openai-cost-url
      :cutoff-date "2023-04")
     (gpt-3.5-turbo
      :description "More expensive & less capable than GPT-4o-mini; use that instead"
@@ -631,6 +647,7 @@ the same as t."
      :context-window 16.358
      :input-cost 0.50
      :output-cost 1.50
+     :cost-url ,gptel--openai-cost-url
      :cutoff-date "2021-09")
     (gpt-3.5-turbo-16k
      :description "More expensive & less capable than GPT-4o-mini; use that instead"
@@ -638,6 +655,7 @@ the same as t."
      :context-window 16.385
      :input-cost 3
      :output-cost 4
+     :cost-url ,gptel--openai-cost-url
      :cutoff-date "2021-09"))
   "List of available OpenAI models and associated properties.
 Keys:
@@ -654,6 +672,8 @@ Keys:
 
 - `:output-cost': the output cost, in US dollars per million tokens.
 
+- `:cost-url': web page for current model pricing.
+
 - `:cutoff-date': the knowledge cutoff date.
 
 - `:request-params': a plist of additional request parameters to
@@ -1442,6 +1462,78 @@ file."
 
 (declare-function gptel-context--wrap "gptel-context")
 
+;; See (info "(elisp) Emacs Lisp Coding Conventions") for justification
+;; for not prefixing this function name with `gptel-'.
+(defun describe-gptel-model (backend model)
+  "Display a description of a gptel model MODEL of BACKEND."
+  (interactive
+   (nthcdr 1
+	   (cl-loop
+	    with default = (concat (gptel-backend-name gptel-backend) ":"
+				   (gptel--model-name gptel-model))
+	    for (name . backend) in gptel--known-backends
+	    nconc (cl-loop for model in (gptel-backend-models backend)
+			   collect (list (concat name ":"
+						 (gptel--model-name model))
+					 backend model))
+	    into models-alist
+	    finally return (assoc (completing-read
+				   (format-prompt "Describe gptel model" default)
+				   models-alist nil t nil nil
+				   default) models-alist))))
+  (help-setup-xref (list #'describe-gptel-model backend model)
+		   (called-interactively-p 'interactive))
+  (with-help-window (help-buffer)
+    (with-current-buffer standard-output
+      (cl-flet ((sect (name val)
+		      (insert (make-string (max 0
+						(- 18 (string-width name)))
+					   ?\s)
+			      (concat name ": "))
+		      (insert (if (listp val)
+				  (string-join (mapcar
+						#'(lambda (s)
+						    (if (symbolp s)
+							(symbol-name s)
+						      s))
+						val) ", ")
+				val))
+		      (insert "\n")))
+	(let* ((stream (gptel-backend-stream backend))
+	       (desc (get model :description))
+	       (caps (get model :capabilities))
+	       (context (get model :context-window))
+	       (maxout (get model :max-output))
+	       (cost-url (get model :cost-url))
+	       (cutoff (get model :cutoff-date))
+	       (mime-types (get model :mime-types))
+	       (input-cost (get model :input-cost))
+	       (output-cost (get model :output-cost)))
+	  (insert (format "%s:%s is a gptel Large Language Model (LLM).\n\n"
+			  (gptel-backend-name backend)
+			  (gptel--model-name model)))
+	  (when desc (sect "Description" desc))
+	  (sect "Backend endpoint"
+		(format "%s://%s%s"
+			(gptel-backend-protocol backend)
+			(gptel-backend-host backend)
+			(gptel-backend-endpoint backend)))
+	  (sect "Streaming" (if stream "yes" "no"))
+	  (when caps (sect "Capabilities" caps))
+	  (when context (sect "Context window" (format "%dk" context)))
+	  (when maxout (sect "Maximum output" (format "%d tokens" maxout)))
+	  (when cutoff (sect "Cut-off date" cutoff))
+	  (when mime-types (sect "MIME types" mime-types))
+	  (when input-cost (sect "Input cost"
+				 (format "$%6.2f per 1M tokens" input-cost)))
+	  (when output-cost (sect "Output cost"
+				  (format "$%6.2f per 1M tokens" output-cost)))
+	  (when (or input-cost output-cost)
+	    (insert (format "%20sPrices subject to change.\n" " "))
+	    (when cost-url
+	      (insert (make-string 20 ?\s))
+	      (help-insert-xref-button cost-url 'help-url cost-url))))))))
+
 
 ;;; Tool use