Release v0.4.0 - Google Gemini Provider

Major Features:
- Full Google Gemini provider support with native batch API
- 50% cost savings through Google's batch processing
- Support for 5 Gemini models (gemini-2.5-flash, gemini-2.5-pro, etc.)
- Structured output and file processing support
- Token counting using Google's official API

Improvements:
- Better exception handling with specific error types
- File path sanitization for batch IDs with slashes
- Cost estimation follows Anthropic pattern (returns 0 when tokencost unavailable)
- Added google-genai dependency for proper CI/CD support
- Updated documentation and README

Bug Fixes:
- Fixed JSON serialization issues with GenerateContentResponse
- Fixed DOCX file type incorrectly listed for Gemini
- Fixed CI failures due to missing google-genai dependency
- Fixed batch ID sanitization for file saving

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>

Author: agamm

README.md

@@ -6,11 +6,11 @@ Unified Python API for AI Batch requests with cost tracking, Pydantic responses,
 
 <img width="1328" height="598" alt="image" src="https://github.com/user-attachments/assets/b1b43070-f664-41a2-a85d-e2c589df556c" />
 
-*This library is currently in alpha - so there will be breaking changes*
+*This library is currently in beta - so there will be breaking changes*
 
 ## Why AI-batching?
 
-AI providers offer batch APIs that process requests asynchronously at 50% reduced cost compared to real-time APIs. This is ideal for workloads like document processing, data analysis, and content generation where immediate responses aren't required. However, managing batch jobs across providers, tracking costs, handling failures, and mapping citations back to source documents quickly becomes complex - that's where Batchata comes in.
+AI providers offer batch APIs that process requests asynchronously at 50% reduced cost compared to real-time APIs. This is ideal for offline or batch processing tasks. However, managing batch jobs across providers, tracking costs, handling failures, and mapping citations back to source documents quickly becomes complex - that's where Batchata comes in.
 
 ## Batchata Features
 

docs/batchata.html

@@ -3143,13 +3143,13 @@ <h2 id="configuration">Configuration</h2>
 </span><span id="Job-13"><a href="#Job-13"><span class="linenos">13</span></a><span class="k">class</span><span class="w"> </span><span class="nc">Job</span><span class="p">:</span>
 </span><span id="Job-14"><a href="#Job-14"><span class="linenos">14</span></a><span class="w">    </span><span class="sd">&quot;&quot;&quot;Configuration for a single AI job.</span>
 </span><span id="Job-15"><a href="#Job-15"><span class="linenos">15</span></a><span class="sd">    </span>
-</span><span id="Job-16"><a href="#Job-16"><span class="linenos">16</span></a><span class="sd">    Either provide messages OR file+prompt, not both.</span>
+</span><span id="Job-16"><a href="#Job-16"><span class="linenos">16</span></a><span class="sd">    Either provide messages OR prompt (with optional file), not both.</span>
 </span><span id="Job-17"><a href="#Job-17"><span class="linenos">17</span></a><span class="sd">    </span>
 </span><span id="Job-18"><a href="#Job-18"><span class="linenos">18</span></a><span class="sd">    Attributes:</span>
 </span><span id="Job-19"><a href="#Job-19"><span class="linenos">19</span></a><span class="sd">        id: Unique identifier for the job</span>
 </span><span id="Job-20"><a href="#Job-20"><span class="linenos">20</span></a><span class="sd">        messages: Chat messages for direct message input</span>
-</span><span id="Job-21"><a href="#Job-21"><span class="linenos">21</span></a><span class="sd">        file: File path for file-based input</span>
-</span><span id="Job-22"><a href="#Job-22"><span class="linenos">22</span></a><span class="sd">        prompt: Prompt to use with file input</span>
+</span><span id="Job-21"><a href="#Job-21"><span class="linenos">21</span></a><span class="sd">        file: Optional file path for file-based input</span>
+</span><span id="Job-22"><a href="#Job-22"><span class="linenos">22</span></a><span class="sd">        prompt: Prompt text (can be used alone or with file)</span>
 </span><span id="Job-23"><a href="#Job-23"><span class="linenos">23</span></a><span class="sd">        model: Model name (e.g., &quot;claude-3-sonnet&quot;)</span>
 </span><span id="Job-24"><a href="#Job-24"><span class="linenos">24</span></a><span class="sd">        temperature: Sampling temperature (0.0-1.0)</span>
 </span><span id="Job-25"><a href="#Job-25"><span class="linenos">25</span></a><span class="sd">        max_tokens: Maximum tokens to generate</span>
@@ -3175,8 +3175,8 @@ <h2 id="configuration">Configuration</h2>
 </span><span id="Job-45"><a href="#Job-45"><span class="linenos">45</span></a>        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">file</span> <span class="ow">and</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">prompt</span><span class="p">:</span>
 </span><span id="Job-46"><a href="#Job-46"><span class="linenos">46</span></a>            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;File input requires a prompt&quot;</span><span class="p">)</span>
 </span><span id="Job-47"><a href="#Job-47"><span class="linenos">47</span></a>        
-</span><span id="Job-48"><a href="#Job-48"><span class="linenos">48</span></a>        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">messages</span> <span class="ow">and</span> <span class="ow">not</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">file</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">prompt</span><span class="p">):</span>
-</span><span id="Job-49"><a href="#Job-49"><span class="linenos">49</span></a>            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Must provide either messages or file+prompt&quot;</span><span class="p">)</span>
+</span><span id="Job-48"><a href="#Job-48"><span class="linenos">48</span></a>        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">messages</span> <span class="ow">and</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">prompt</span><span class="p">:</span>
+</span><span id="Job-49"><a href="#Job-49"><span class="linenos">49</span></a>            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Must provide either messages or prompt&quot;</span><span class="p">)</span>
 </span><span id="Job-50"><a href="#Job-50"><span class="linenos">50</span></a>    
 </span><span id="Job-51"><a href="#Job-51"><span class="linenos">51</span></a>    <span class="k">def</span><span class="w"> </span><span class="nf">to_dict</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
 </span><span id="Job-52"><a href="#Job-52"><span class="linenos">52</span></a><span class="w">        </span><span class="sd">&quot;&quot;&quot;Serialize for state persistence.&quot;&quot;&quot;</span>
@@ -3218,13 +3218,13 @@ <h2 id="configuration">Configuration</h2>
 
             <div class="docstring"><p>Configuration for a single AI job.</p>
 
-<p>Either provide messages OR file+prompt, not both.</p>
+<p>Either provide messages OR prompt (with optional file), not both.</p>
 
 <p>Attributes:
     id: Unique identifier for the job
     messages: Chat messages for direct message input
-    file: File path for file-based input
-    prompt: Prompt to use with file input
+    file: Optional file path for file-based input
+    prompt: Prompt text (can be used alone or with file)
     model: Model name (e.g., "claude-3-sonnet")
     temperature: Sampling temperature (0.0-1.0)
     max_tokens: Maximum tokens to generate