From fb70d7f7eb9f7971fc157ef6597a4c18f4603a5b Mon Sep 17 00:00:00 2001
From: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Date: Thu, 16 Jul 2026 12:08:38 +0000
Subject: [PATCH] feat(elevenlabs): add scribe_v2 to model cost map

---
 .../model_prices_and_context_window_backup.json   | 15 +++++++++++++++
 model_prices_and_context_window.json              | 15 +++++++++++++++
 tests/llm_translation/test_elevenlabs.py          |  9 +++++++++
 3 files changed, 39 insertions(+)

diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json
index dedb9bbf40a..d08dfb57c0c 100644
--- a/litellm/model_prices_and_context_window_backup.json
+++ b/litellm/model_prices_and_context_window_backup.json
@@ -15353,6 +15353,21 @@
             "/v1/audio/transcriptions"
         ]
     },
+    "elevenlabs/scribe_v2": {
+        "input_cost_per_second": 6.11e-05,
+        "litellm_provider": "elevenlabs",
+        "metadata": {
+            "calculation": "$0.22/hour = $0.00366/minute = $0.0000611 per second",
+            "notes": "ElevenLabs Scribe v2 - batch speech-to-text model released January 2026, lowest word error rate on industry benchmarks with 99 language support",
+            "original_pricing_per_hour": 0.22
+        },
+        "mode": "audio_transcription",
+        "output_cost_per_second": 0.0,
+        "source": "https://elevenlabs.io/pricing/api",
+        "supported_endpoints": [
+            "/v1/audio/transcriptions"
+        ]
+    },
     "elevenlabs/scribe_v1_experimental": {
         "input_cost_per_second": 6.11e-05,
         "litellm_provider": "elevenlabs",
diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json
index e10dde793d1..a07cee8a780 100644
--- a/model_prices_and_context_window.json
+++ b/model_prices_and_context_window.json
@@ -15353,6 +15353,21 @@
             "/v1/audio/transcriptions"
         ]
     },
+    "elevenlabs/scribe_v2": {
+        "input_cost_per_second": 6.11e-05,
+        "litellm_provider": "elevenlabs",
+        "metadata": {
+            "calculation": "$0.22/hour = $0.00366/minute = $0.0000611 per second",
+            "notes": "ElevenLabs Scribe v2 - batch speech-to-text model released January 2026, lowest word error rate on industry benchmarks with 99 language support",
+            "original_pricing_per_hour": 0.22
+        },
+        "mode": "audio_transcription",
+        "output_cost_per_second": 0.0,
+        "source": "https://elevenlabs.io/pricing/api",
+        "supported_endpoints": [
+            "/v1/audio/transcriptions"
+        ]
+    },
     "elevenlabs/scribe_v1_experimental": {
         "input_cost_per_second": 6.11e-05,
         "litellm_provider": "elevenlabs",
diff --git a/tests/llm_translation/test_elevenlabs.py b/tests/llm_translation/test_elevenlabs.py
index b6c838d2300..47f61c749ee 100644
--- a/tests/llm_translation/test_elevenlabs.py
+++ b/tests/llm_translation/test_elevenlabs.py
@@ -213,3 +213,12 @@ class TestElevenLabsTextToSpeechTransformation:
 
         assert voice_id in url
         assert "output_format=pcm_44100" in url
+
+
+class TestElevenLabsScribeV2Pricing:
+    def test_scribe_v2_model_info(self):
+        info = litellm.get_model_info("elevenlabs/scribe_v2")
+        assert info["litellm_provider"] == "elevenlabs"
+        assert info["mode"] == "audio_transcription"
+        assert info["input_cost_per_second"] == pytest.approx(0.22 / 3600, rel=1e-3)
+        assert info["output_cost_per_second"] == 0.0