diff --git a/aider/models.py b/aider/models.py
index 97575993b..6e91d6aab 100644
--- a/aider/models.py
+++ b/aider/models.py
@@ -841,7 +841,7 @@ MODEL_SETTINGS = [
use_repo_map=True,
streaming=False,
use_temperature=False,
- # extra_params=dict(extra_body=dict(reasoning_effort="high")),
+ extra_params=dict(extra_body=dict(reasoning_effort="high")),
),
ModelSettings(
"openrouter/qwen/qwen-2.5-coder-32b-instruct",
diff --git a/aider/website/_data/r1_architect.yml b/aider/website/_data/r1_architect.yml
new file mode 100644
index 000000000..c036c7de8
--- /dev/null
+++ b/aider/website/_data/r1_architect.yml
@@ -0,0 +1,138 @@
+
+
+
+- dirname: 2025-01-23-19-14-48--r1-architect-sonnet
+ test_cases: 225
+ model: R1+Sonnet
+ edit_format: architect
+ commit_hash: 05a77c7
+ editor_model: claude-3-5-sonnet-20241022
+ editor_edit_format: editor-diff
+ pass_rate_1: 27.1
+ pass_rate_2: 64.0
+ pass_num_1: 61
+ pass_num_2: 144
+ percent_cases_well_formed: 100.0
+ error_outputs: 2
+ num_malformed_responses: 0
+ num_with_malformed_responses: 0
+ user_asks: 392
+ lazy_comments: 6
+ syntax_errors: 0
+ indentation_errors: 0
+ exhausted_context_windows: 0
+ test_timeouts: 5
+ total_tests: 225
+ command: aider --model deepseek/deepseek-reasoner
+ date: 2025-01-23
+ versions: 0.72.3.dev
+ seconds_per_case: 251.6
+ total_cost: 13.2933
+
+- dirname: 2025-01-20-19-11-38--ds-turns-upd-cur-msgs-fix-with-summarizer
+ test_cases: 225
+ model: R1
+ edit_format: diff
+ commit_hash: 5650697-dirty
+ pass_rate_1: 26.7
+ pass_rate_2: 56.9
+ pass_num_1: 60
+ pass_num_2: 128
+ percent_cases_well_formed: 96.9
+ error_outputs: 8
+ num_malformed_responses: 7
+ num_with_malformed_responses: 7
+ user_asks: 15
+ lazy_comments: 0
+ syntax_errors: 0
+ indentation_errors: 0
+ exhausted_context_windows: 1
+ test_timeouts: 5
+ total_tests: 225
+ command: aider --model deepseek/deepseek-reasoner
+ date: 2025-01-20
+ versions: 0.71.2.dev
+ seconds_per_case: 113.7
+ total_cost: 5.4193
+
+
+- dirname: 2024-12-21-19-23-03--polyglot-o1-hard-diff
+ test_cases: 224
+ model: o1
+ edit_format: diff
+ commit_hash: a755079-dirty
+ pass_rate_1: 23.7
+ pass_rate_2: 61.7
+ pass_num_1: 53
+ pass_num_2: 139
+ percent_cases_well_formed: 91.5
+ error_outputs: 25
+ num_malformed_responses: 24
+ num_with_malformed_responses: 19
+ user_asks: 16
+ lazy_comments: 0
+ syntax_errors: 0
+ indentation_errors: 0
+ exhausted_context_windows: 0
+ test_timeouts: 2
+ total_tests: 225
+ command: aider --model openrouter/openai/o1
+ date: 2024-12-21
+ versions: 0.69.2.dev
+ seconds_per_case: 133.2
+ total_cost: 186.4958
+
+
+- dirname: 2024-12-25-13-31-51--deepseekv3preview-diff2
+ test_cases: 225
+ model: DeepSeek V3
+ edit_format: diff
+ commit_hash: 0a23c4a-dirty
+ pass_rate_1: 22.7
+ pass_rate_2: 48.4
+ pass_num_1: 51
+ pass_num_2: 109
+ percent_cases_well_formed: 98.7
+ error_outputs: 7
+ num_malformed_responses: 7
+ num_with_malformed_responses: 3
+ user_asks: 19
+ lazy_comments: 0
+ syntax_errors: 0
+ indentation_errors: 0
+ exhausted_context_windows: 0
+ test_timeouts: 8
+ total_tests: 225
+ command: aider --model deepseek/deepseek-chat
+ date: 2024-12-25
+ versions: 0.69.2.dev
+ seconds_per_case: 34.8
+ total_cost: 0.3369
+
+
+
+- dirname: 2025-01-17-19-44-33--sonnet-baseline-jan-17
+ test_cases: 225
+ model: Sonnet
+ edit_format: diff
+ commit_hash: 6451d59
+ pass_rate_1: 22.2
+ pass_rate_2: 51.6
+ pass_num_1: 50
+ pass_num_2: 116
+ percent_cases_well_formed: 99.6
+ error_outputs: 2
+ num_malformed_responses: 1
+ num_with_malformed_responses: 1
+ user_asks: 11
+ lazy_comments: 0
+ syntax_errors: 0
+ indentation_errors: 0
+ exhausted_context_windows: 1
+ test_timeouts: 8
+ total_tests: 225
+ command: aider --model claude-3-5-sonnet-20241022
+ date: 2025-01-17
+ versions: 0.71.2.dev
+ seconds_per_case: 21.4
+ total_cost: 14.4063
diff --git a/aider/website/_posts/2025-01-23-r1.md b/aider/website/_posts/2025-01-23-r1.md
index 1198cc8c5..264887e33 100644
--- a/aider/website/_posts/2025-01-23-r1.md
+++ b/aider/website/_posts/2025-01-23-r1.md
@@ -1,5 +1,5 @@
---
-title: r1 tops aider's polyglot leaderboard
+title: R1+Sonnet set SOTA on aider's polyglot benchmark
#excerpt: o1 scores the top result on aider's new multi-language, more challenging coding benchmark.
#highlight_image: /assets/o1-polyglot.jpg
draft: false
@@ -9,12 +9,24 @@ nav_exclude: true
{{ page.date | date: "%B %d, %Y" }}
{% endif %}
-# r1 tops aider's polyglot leaderboard
+# R1+Sonnet set SOTA on aider's polyglot benchmark
{: .no_toc }
+Aider supports using a pair of models for coding:
+- An Architect model is asked to describe how to solve the coding problem. Thinking/reasoning models often work well in this role.
+- An Editor model is given the Architect's solution and asked to produce specific code editing instructions to apply those changes to existing source files.
+
+**R1 as architect with Sonnet as editor has set a new SOTA of 64.0%** on the
+[aider polyglot benchmark](/2024/12/21/polyglot.html).
+They achieve this at **14X less cost** compared to the previous o1 SOTA result.
+
+Using o1 or R1 as architect with various other editor models didn't produce significantly
+better results than using them alone.
+This is in contrast to the first wave of thinking models like o1-preview and o1-mini,
+which improved when paired with many different editor models.
## Results