From 559a4cde23863af48b11c0bb1605dec26dab4b80 Mon Sep 17 00:00:00 2001
From: Gregory Nwosu <193151+gregnwosu@users.noreply.github.com>
Date: Mon, 8 Apr 2024 02:03:54 +0100
Subject: [PATCH] created defaults for response["eval_count"]

there is no way in litellm to disable the cache in ollama that is removing the eval_count response keys from the json.
This PR allows the code to create sensible defaults for when the response is empty
see
- https://github.com/ollama/ollama/issues/1573
- https://github.com/ollama/ollama/issues/2023
---
 litellm/llms/ollama.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/litellm/llms/ollama.py b/litellm/llms/ollama.py
index 779896abfdd..a14c3cb5031 100644
--- a/litellm/llms/ollama.py
+++ b/litellm/llms/ollama.py
@@ -229,7 +229,7 @@ def get_ollama_response(
     model_response["created"] = int(time.time())
     model_response["model"] = "ollama/" + model
     prompt_tokens = response_json.get("prompt_eval_count", len(encoding.encode(prompt)))  # type: ignore
-    completion_tokens = response_json["eval_count"]
+    completion_tokens = response_json.get("eval_count", len(response_json.get("message",dict()).get("content", "")))
     model_response["usage"] = litellm.Usage(
         prompt_tokens=prompt_tokens,
         completion_tokens=completion_tokens,
@@ -331,7 +331,7 @@ async def ollama_acompletion(url, data, model_response, encoding, logging_obj):
             model_response["created"] = int(time.time())
             model_response["model"] = "ollama/" + data["model"]
             prompt_tokens = response_json.get("prompt_eval_count", len(encoding.encode(data["prompt"])))  # type: ignore
-            completion_tokens = response_json["eval_count"]
+            completion_tokens = response_json.get("eval_count", len(response_json.get("message",dict()).get("content", "")))
             model_response["usage"] = litellm.Usage(
                 prompt_tokens=prompt_tokens,
                 completion_tokens=completion_tokens,