Add Copilot SWE to supported models (#58359)

Co-authored-by: Sunbrye Ly <56200261+sunbrye@users.noreply.github.com>

Author: nhu-do

content/copilot/reference/ai-models/model-comparison.md

@@ -42,6 +42,7 @@ Use this table to find a suitable model quickly, see more detail in the sections
 | {% data variables.copilot.copilot_gemini_25_pro %}    | Deep reasoning and debugging                     | Complex code generation, debugging, and research workflows              | Reasoning, vision             | [{% data variables.copilot.copilot_gemini_25_pro %} model card](https://storage.googleapis.com/model-cards/documents/gemini-2.5-pro.pdf)                                    |
 | {% data variables.copilot.copilot_grok_code %}        | General-purpose coding and writing               | Fast, accurate code completions and explanations                        | Agent mode                    | [{% data variables.copilot.copilot_grok_code %} model card](https://data.x.ai/2025-08-20-grok-4-model-card.pdf)                                                             |
 | {% data variables.copilot.copilot_qwen_25 %}          | General-purpose coding and writing               | Code generation, reasoning, and code repair / debugging                 | Reasoning                     | [{% data variables.copilot.copilot_qwen_25 %} model card](https://arxiv.org/pdf/2409.12186)                                                                                 |
+| {% data variables.copilot.copilot_swe_mini %}          | General-purpose coding and writing               | Fast, accurate code completions and explanations                 | Agent mode                     | Coming soon                                                                                |
 
 ## Task: General-purpose coding and writing
 
@@ -52,6 +53,7 @@ Use these models for common development tasks that require a balance of quality,
 | {% data variables.copilot.copilot_gpt_5_codex %}      | Delivers higher-quality code on complex engineering tasks like features, tests, debugging, refactors, and reviews without lengthy instructions. |
 | {% data variables.copilot.copilot_gpt_5_mini %}       | Reliable default for most coding and writing tasks. Fast, accurate, and works well across languages and frameworks.                             |
 | {% data variables.copilot.copilot_grok_code %}        | Specialized for coding tasks. Performs well on code generation, and debugging across multiple languages.                                        |
+| {% data variables.copilot.copilot_swe_mini %}        | Specialized for fast, accurate code completions and explanations.                                        |
 
 ### When to use these models
 

content/copilot/reference/ai-models/model-hosting.md

@@ -31,6 +31,14 @@ Used for:
 
 When using OpenAI's models, input requests and output responses continue to run through {% data variables.product.prodname_copilot %}'s content filters for public code matching, when applied, along with those for harmful or offensive content.
 
+## OpenAI models fine-tuned by Microsoft
+
+Used for:
+
+* {% data variables.copilot.copilot_swe_mini %}
+
+{% data variables.copilot.copilot_swe_mini %} is deployed on {% data variables.product.github %} managed Azure OpenAI tenant.
+
 ## Anthropic models
 
 Used for:

data/reusables/copilot/model-use-cases/copilot-swe.md

@@ -0,0 +1 @@
+{% data variables.copilot.copilot_swe_mini %} is a good choice for common development tasks that benefit from general-purpose coding and writing. If you're working on tasks that require fast iteration, accurate code completions and explanations, {% data variables.copilot.copilot_swe_mini %} is likely the model to use.

data/tables/copilot/model-multipliers.yml

@@ -52,3 +52,7 @@
 - name: Grok Code Fast 1
   multiplier_paid: 0.25
   multiplier_free: Not applicable
+
+- name: Copilot-SWE mini
+  multiplier_paid: Not applicable
+  multiplier_free: 1

data/tables/copilot/model-release-status.yml

@@ -91,3 +91,11 @@
   agent_mode: true
   ask_mode: true
   edit_mode: true
+
+# Fine-tuned OAI models
+- name: 'Copilot-SWE mini'
+  provider: 'Fine-tuned GPT-5 mini'
+  release_status: 'Public preview'
+  agent_mode: true
+  ask_mode: true
+  edit_mode: true

data/tables/copilot/model-supported-clients.yml

@@ -92,3 +92,11 @@
   eclipse: true
   xcode: true
   jetbrains: true
+
+- name: Copilot-SWE mini
+  dotcom: false
+  vscode: true
+  vs: false
+  eclipse: false
+  xcode: false
+  jetbrains: false

data/tables/copilot/model-supported-plans.yml

@@ -81,3 +81,10 @@
   pro_plus: true
   business: true
   enterprise: true
+
+- name: Copilot-SWE mini
+  free: true
+  pro: false
+  pro_plus: false
+  business: false
+  enterprise: false

data/variables/copilot.yml

@@ -154,6 +154,8 @@ copilot_o4_mini: 'o4-mini'
 copilot_grok_code: 'Grok Code Fast 1'
 # Qwen:
 copilot_qwen_25: 'Qwen2.5'
+# Microsoft fine-tuned:
+copilot_swe_mini: 'Copilot-SWE mini'
 
 ## Current model used by Copilot coding agent and Copilot CLI
 cca_current_model: '{% data variables.copilot.copilot_claude_sonnet_40 %}'