feat: together now supports base64 embedding encoding

recordings for: ./scripts/integration-tests.sh --stack-config server:ci-tests --suite base --setup together --subdirs inference --pattern openai

tests/integration/recordings/responses/6841bb14fa8d.json

@@ -0,0 +1,61 @@
+{
+  "request": {
+    "method": "POST",
+    "url": "https://api.together.xyz/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "meta-llama/Llama-3.3-70B-Instruct-Turbo-Free",
+      "messages": [
+        {
+          "role": "user",
+          "content": "Test trace openai with temperature 0"
+        }
+      ],
+      "max_tokens": 100,
+      "stream": false,
+      "temperature": 0.7
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "meta-llama/Llama-3.3-70B-Instruct-Turbo-Free"
+  },
+  "response": {
+    "body": {
+      "__type__": "openai.types.chat.chat_completion.ChatCompletion",
+      "__data__": {
+        "id": "oCfxEyX-4Yz4kd-984c2b58fd3f4d13",
+        "choices": [
+          {
+            "finish_reason": "length",
+            "index": 0,
+            "logprobs": null,
+            "message": {
+              "content": "When using a language model like OpenAI with a temperature of 0, the model is essentially forced to produce the most likely next word in a sequence, given the context. This means that the output will be very deterministic and less diverse, as the model is not allowed to explore less likely options.\n\nHere's an example of how this could work in practice:\n\n**Prompt:** Write a short story about a character who discovers a hidden world.\n\n**Temperature 0 Response:**\nIn a small village nestled",
+              "refusal": null,
+              "role": "assistant",
+              "annotations": null,
+              "audio": null,
+              "function_call": null,
+              "tool_calls": []
+            },
+            "seed": 9269366008132817000
+          }
+        ],
+        "created": 1758820586,
+        "model": "meta-llama/Llama-3.3-70B-Instruct-Turbo-Free",
+        "object": "chat.completion",
+        "service_tier": null,
+        "system_fingerprint": null,
+        "usage": {
+          "completion_tokens": 100,
+          "prompt_tokens": 43,
+          "total_tokens": 143,
+          "completion_tokens_details": null,
+          "prompt_tokens_details": null,
+          "cached_tokens": 0
+        },
+        "prompt": []
+      }
+    },
+    "is_streaming": false
+  }
+}