From 426ec39816fc9f0da4fecf10d6b0492cc1308fe5 Mon Sep 17 00:00:00 2001
From: John Mangel <john@inpractice.ai>
Date: Thu, 2 Jul 2026 22:00:37 +0000
Subject: [PATCH] Fix Bedrock Converse input_tokens double-subtracting cache
 tokens

AWS Bedrock reports inputTokens as already non-cached; cacheReadInputTokens
and cacheWriteInputTokens are separate buckets, not folded in. Subtracting
them (as inclusive providers like OpenAI/Gemini require) understated input
and floored it to zero on cache-heavy turns. Pass inputTokens through, as
the Anthropic provider already does for the same Claude models.

Updates the converse chat spec: the prior test encoded the subtraction on a
payload that can't occur (inputTokens already excludes cache per AWS), plus a
regression for the floor-to-zero case.
---
 lib/ruby_llm/protocols/converse/chat.rb       |  9 +++---
 spec/ruby_llm/protocols/converse/chat_spec.rb | 30 +++++++++++++++++--
 2 files changed, 33 insertions(+), 6 deletions(-)

diff --git a/lib/ruby_llm/protocols/converse/chat.rb b/lib/ruby_llm/protocols/converse/chat.rb
index c3cdbe493..46708c37e 100644
--- a/lib/ruby_llm/protocols/converse/chat.rb
+++ b/lib/ruby_llm/protocols/converse/chat.rb
@@ -90,10 +90,11 @@ def parse_completion_body(data, raw:)
         end
 
         def input_tokens(usage)
-          input_tokens = usage['inputTokens']
-          return unless input_tokens
-
-          [input_tokens.to_i - usage['cacheReadInputTokens'].to_i - usage['cacheWriteInputTokens'].to_i, 0].max
+          # AWS Bedrock reports inputTokens as already non-cached; cacheReadInputTokens and
+          # cacheWriteInputTokens are separate buckets, not folded into inputTokens. Subtracting
+          # them (as inclusive providers require) understates input and floors to zero on cache
+          # hits. See https://docs.aws.amazon.com/bedrock/latest/userguide/prompt-caching.html
+          usage['inputTokens']
         end
 
         def reasoning_tokens(usage)
diff --git a/spec/ruby_llm/protocols/converse/chat_spec.rb b/spec/ruby_llm/protocols/converse/chat_spec.rb
index 7e98a1810..6d845073d 100644
--- a/spec/ruby_llm/protocols/converse/chat_spec.rb
+++ b/spec/ruby_llm/protocols/converse/chat_spec.rb
@@ -4,7 +4,9 @@
 
 RSpec.describe RubyLLM::Protocols::Converse::Chat do
   describe '.parse_completion_response' do
-    it 'normalizes cache read and write tokens out of input tokens' do
+    it 'exposes AWS inputTokens as-is (already non-cached) and keeps cache buckets separate' do
+      # Per AWS, inputTokens already excludes cache; a real payload sends the non-cached count
+      # directly, with cache read/write reported separately.
       response_body = {
         'modelId' => 'anthropic.claude-sonnet-4-5-20250929-v1:0',
         'output' => {
@@ -13,7 +15,7 @@
           }
         },
         'usage' => {
-          'inputTokens' => 100,
+          'inputTokens' => 50,
           'outputTokens' => 5,
           'cacheReadInputTokens' => 40,
           'cacheWriteInputTokens' => 10
@@ -29,6 +31,30 @@
       expect(message.cache_creation_tokens).to eq(10)
     end
 
+    it 'does not subtract cache buckets or floor to zero when the cached prefix exceeds fresh input' do
+      response_body = {
+        'modelId' => 'anthropic.claude-sonnet-4-5-20250929-v1:0',
+        'output' => {
+          'message' => {
+            'content' => [{ 'text' => 'Hi!' }]
+          }
+        },
+        'usage' => {
+          'inputTokens' => 3,
+          'outputTokens' => 5,
+          'cacheReadInputTokens' => 7714,
+          'cacheWriteInputTokens' => 327
+        }
+      }
+
+      response = instance_double(Faraday::Response, body: response_body)
+      message = described_class.parse_completion_response(response)
+
+      expect(message.input_tokens).to eq(3)
+      expect(message.cached_tokens).to eq(7714)
+      expect(message.cache_creation_tokens).to eq(327)
+    end
+
     it 'preserves raw stopReason as finish_reason' do
       response_body = {
         'modelId' => 'amazon.nova-lite-v1:0',