From 51cba2e6a4d5bcf6b73faf7333ee7a22063d7564 Mon Sep 17 00:00:00 2001
From: Aiden Cline <rekram1-node@users.noreply.github.com>
Date: Thu, 13 Aug 2026 17:16:31 +0000
Subject: [PATCH 1/2] feat: add Gemini 3.7 Flash

---
 models/google/gemini-3.7-flash.toml           | 68 +++++++++++++++++++
 .../models/gemini-3.7-flash.toml              | 12 ++++
 providers/google/models/gemini-3.7-flash.toml | 15 ++++
 3 files changed, 95 insertions(+)
 create mode 100644 models/google/gemini-3.7-flash.toml
 create mode 100644 providers/google-vertex/models/gemini-3.7-flash.toml
 create mode 100644 providers/google/models/gemini-3.7-flash.toml

diff --git a/models/google/gemini-3.7-flash.toml b/models/google/gemini-3.7-flash.toml
new file mode 100644
index 0000000000..044c4a2efe
--- /dev/null
+++ b/models/google/gemini-3.7-flash.toml
@@ -0,0 +1,68 @@
+name = "Gemini 3.7 Flash"
+description = "High-efficiency Gemini model for agentic workflows, coding, and multimodal reasoning"
+family = "gemini-flash"
+release_date = "2026-08-13"
+last_updated = "2026-08-13"
+attachment = true
+reasoning = true
+temperature = true
+tool_call = true
+structured_output = true
+knowledge = "2026-03"
+open_weights = false
+
+[limit]
+context = 1_048_576
+output = 65_536
+
+[modalities]
+input = ["text", "image", "video", "audio", "pdf"]
+output = ["text"]
+
+[[benchmarks]]
+name = "FrontierCode"
+score = 43.6
+metric = "score"
+version = "1.1 Main"
+source = "https://deepmind.google/models/model-cards/gemini-3-7-flash/"
+date = "2026-08-13"
+
+[[benchmarks]]
+name = "DeepSWE"
+score = 65.3
+metric = "resolve rate"
+version = "1.1"
+source = "https://deepmind.google/models/model-cards/gemini-3-7-flash/"
+date = "2026-08-13"
+
+[[benchmarks]]
+name = "Terminal-Bench"
+score = 85.8
+metric = "accuracy"
+version = "2.1"
+source = "https://deepmind.google/models/model-cards/gemini-3-7-flash/"
+date = "2026-08-13"
+
+[[benchmarks]]
+name = "AutomationBench"
+score = 30.4
+metric = "accuracy"
+dataset = "private set"
+source = "https://deepmind.google/models/model-cards/gemini-3-7-flash/"
+date = "2026-08-13"
+
+[[benchmarks]]
+name = "GDP.pdf"
+score = 34.0
+metric = "accuracy"
+source = "https://deepmind.google/models/model-cards/gemini-3-7-flash/"
+date = "2026-08-13"
+
+[[benchmarks]]
+name = "GDM-MRCR"
+score = 97.0
+metric = "accuracy"
+variant = "128k average, 8-needle"
+version = "v2"
+source = "https://deepmind.google/models/model-cards/gemini-3-7-flash/"
+date = "2026-08-13"
diff --git a/providers/google-vertex/models/gemini-3.7-flash.toml b/providers/google-vertex/models/gemini-3.7-flash.toml
new file mode 100644
index 0000000000..fe4bb315ca
--- /dev/null
+++ b/providers/google-vertex/models/gemini-3.7-flash.toml
@@ -0,0 +1,12 @@
+# Sources:
+# - https://cloud.google.com/gemini-enterprise-agent-platform/generative-ai/pricing
+# - https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-7-flash
+# - https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/thinking
+base_model = "google/gemini-3.7-flash"
+reasoning_options = [{ type = "effort", values = ["low", "medium", "high"] }]
+
+[cost]
+input = 1.5
+output = 7.5
+cache_read = 0.15
+input_audio = 1.5
diff --git a/providers/google/models/gemini-3.7-flash.toml b/providers/google/models/gemini-3.7-flash.toml
new file mode 100644
index 0000000000..1e6abbafa8
--- /dev/null
+++ b/providers/google/models/gemini-3.7-flash.toml
@@ -0,0 +1,15 @@
+# Sources:
+# - https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-gemini-3-7-flash/
+# - https://ai.google.dev/gemini-api/docs/pricing
+# - https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/thinking
+base_model = "google/gemini-3.7-flash"
+
+[[reasoning_options]]
+type = "effort"
+values = ["low", "medium", "high"]
+
+[cost]
+input = 1.50
+output = 7.50
+cache_read = 0.15
+input_audio = 1.50

From f3d8a5837526a88c98ffde94242098b00524269c Mon Sep 17 00:00:00 2001
From: Aiden Cline <rekram1-node@users.noreply.github.com>
Date: Thu, 13 Aug 2026 17:25:08 +0000
Subject: [PATCH 2/2] fix: use Gemini 3.7 introductory pricing

---
 providers/google-vertex/models/gemini-3.7-flash.toml | 8 ++++----
 providers/google/models/gemini-3.7-flash.toml        | 8 ++++----
 2 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/providers/google-vertex/models/gemini-3.7-flash.toml b/providers/google-vertex/models/gemini-3.7-flash.toml
index fe4bb315ca..3da8cc6ab6 100644
--- a/providers/google-vertex/models/gemini-3.7-flash.toml
+++ b/providers/google-vertex/models/gemini-3.7-flash.toml
@@ -6,7 +6,7 @@ base_model = "google/gemini-3.7-flash"
 reasoning_options = [{ type = "effort", values = ["low", "medium", "high"] }]
 
 [cost]
-input = 1.5
-output = 7.5
-cache_read = 0.15
-input_audio = 1.5
+input = 0.75
+output = 3.75
+cache_read = 0.075
+input_audio = 0.75
diff --git a/providers/google/models/gemini-3.7-flash.toml b/providers/google/models/gemini-3.7-flash.toml
index 1e6abbafa8..eb3c50e2e2 100644
--- a/providers/google/models/gemini-3.7-flash.toml
+++ b/providers/google/models/gemini-3.7-flash.toml
@@ -9,7 +9,7 @@ type = "effort"
 values = ["low", "medium", "high"]
 
 [cost]
-input = 1.50
-output = 7.50
-cache_read = 0.15
-input_audio = 1.50
+input = 0.75
+output = 3.75
+cache_read = 0.075
+input_audio = 0.75