synth-ai 0.2.8.dev11__py3-none-any.whl → 0.2.8.dev12__py3-none-any.whl

synth_ai/demos/demo_task_apps/math/modal_task_app.py

@@ -205,12 +205,14 @@ def fastapi_app():
             """Return Hendrycks MATH problem/answer and tool schema for a seed."""
             q, a = _load_hendrycks_problem(int(seed), subject=subject)
             tools = [{
-                "name": "interact",
-                "description": "Submit one or more actions to the math environment.",
+                "name": "submit_answer",
+                "description": "Provide the final numerical or algebraic answer for the current math problem.",
                 "parameters": {
                     "type": "object",
-                    "properties": {"actions": {"type": "array", "items": {"type": "string"}}},
-                    "required": ["actions"],
+                    "properties": {
+                        "answer": {"type": "string", "description": "The proposed final answer"},
+                    },
+                    "required": ["answer"],
                 },
             }]
             return {
@@ -245,7 +247,7 @@ def fastapi_app():
 
     OPENAI_REMOVE_FIELDS = ("stop_after_tool_calls", "thinking_mode", "thinking_budget", "reasoning")
     OPENAI_REMOVE_SAMPLING_FIELDS = ("temperature", "top_p")
-    TOOL_CHOICE_FORCE = {"type": "function", "function": {"name": "interact_many"}}
+    TOOL_CHOICE_FORCE = {"type": "function", "function": {"name": "submit_answer"}}
 
     def _prepare_openai_payload(model: str | None, payload: dict[str, object]) -> dict[str, object]:
         sanitized = dict(payload)
@@ -258,9 +260,9 @@ def fastapi_app():
                 sanitized.pop("max_tokens", None)
             for field in OPENAI_REMOVE_SAMPLING_FIELDS:
                 sanitized.pop(field, None)
-            sanitized["tool_choice"] = TOOL_CHOICE_FORCE
-            sanitized["parallel_tool_calls"] = False
-        return sanitized
+                sanitized["tool_choice"] = TOOL_CHOICE_FORCE
+                sanitized["parallel_tool_calls"] = False
+            return sanitized
 
     @api.post("/proxy/v1/chat/completions")
     def proxy_chat_completions(request: dict[str, object] = Body(...)):
@@ -330,11 +332,11 @@ def fastapi_app():
                     sanitized.pop("max_tokens", None)
                 for field in ("temperature", "top_p"):
                     sanitized.pop(field, None)
-                sanitized["tool_choice"] = {"type": "function", "function": {"name": "interact"}}
+                sanitized["tool_choice"] = {"type": "function", "function": {"name": "submit_answer"}}
                 sanitized["parallel_tool_calls"] = False
             return sanitized
 
-        def _parse_tool_actions(resp: dict[str, Any]) -> list[str]:
+        def _parse_tool_answer(resp: dict[str, Any]) -> str:
             try:
                 choices = resp.get("choices")
                 if isinstance(choices, list) and choices:
@@ -343,7 +345,7 @@ def fastapi_app():
                     if isinstance(tcs, list) and tcs:
                         fn = tcs[0].get("function", {}) if isinstance(tcs[0], dict) else {}
                         args = fn.get("arguments")
-                        obj = {}
+                        obj: dict[str, Any] = {}
                         if isinstance(args, str):
                             try:
                                 obj = _json.loads(args)
@@ -351,12 +353,12 @@ def fastapi_app():
                                 obj = {}
                         elif isinstance(args, dict):
                             obj = args
-                        acts = obj.get("actions")
-                        if isinstance(acts, list):
-                            return [str(a) for a in acts][:5]
+                        ans = obj.get("answer")
+                        if isinstance(ans, str):
+                            return ans.strip()
             except Exception:
                 pass
-            return []
+            return ""
 
         # Single-step rollout: one agent call followed by evaluation of the returned tool answer
         history: list[dict[str, Any]] = []
@@ -373,7 +375,16 @@ def fastapi_app():
             "messages": [{"role": "user", "content": user_prompt}],
             "tools": [{
                 "type": "function",
-                "function": {"name": "interact", "parameters": {"type": "object", "properties": {"actions": {"type": "array", "items": {"type": "string"}}}, "required": ["actions"]}},
+                "function": {
+                    "name": "submit_answer",
+                    "parameters": {
+                        "type": "object",
+                        "properties": {
+                            "answer": {"type": "string"},
+                        },
+                        "required": ["answer"],
+                    },
+                },
             }],
             "max_tokens": 256,
             "temperature": 0.2,
@@ -431,11 +442,11 @@ def fastapi_app():
         except Exception:
             pass
 
-        tool_actions = _parse_tool_actions(data)
-        history.append({"actions": tool_actions})
+        tool_answer = _parse_tool_answer(data)
+        history.append({"answer": tool_answer})
         steps.append({
             "obs": {},
-            "tool_calls": [{"tool_name": "interact", "arguments": _json.dumps({"actions": tool_actions})}],
+            "tool_calls": [{"tool_name": "submit_answer", "arguments": _json.dumps({"answer": tool_answer})}],
             "reward": None,
             "done": False,
             "truncated": False,
@@ -444,13 +455,8 @@ def fastapi_app():
 
         # Evaluate answer correctness using tool output (or fall back to assistant text)
         reward_val = 0.0
-        candidate = ""
+        candidate = tool_answer or ""
         try:
-            if isinstance(tool_actions, list):
-                for s in reversed(tool_actions):
-                    if isinstance(s, str) and s.strip():
-                        candidate = s.strip()
-                        break
             if not candidate and llm_text is not None:
                 candidate = _extract_boxed(llm_text) or llm_text
             if expected_answer is not None:

{synth_ai-0.2.8.dev11.dist-info → synth_ai-0.2.8.dev12.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: synth-ai
-Version: 0.2.8.dev11
+Version: 0.2.8.dev12
 Summary: RL as a service SDK - Core AI functionality and tracing
 Author-email: Synth AI <josh@usesynth.ai>
 License-Expression: MIT

{synth_ai-0.2.8.dev11.dist-info → synth_ai-0.2.8.dev12.dist-info}/RECORD

@@ -30,7 +30,7 @@ synth_ai/demos/demo_task_apps/math/app.py,sha256=gNopoAhwM0vzdKuCa7AwQqSwiV2xagr
 synth_ai/demos/demo_task_apps/math/config.toml,sha256=Kxrzuyj7Az5mvzXaipPIyngKTDqphohf6uSWOHCF5cw,2105
 synth_ai/demos/demo_task_apps/math/deploy_modal.py,sha256=O4745sFuGEZTsygl-mz6ZOFJ7mog8CquXMgMyjFKr_c,2288
 synth_ai/demos/demo_task_apps/math/deploy_task_app.sh,sha256=qVffbAmsiCAxzFDzcxNVF4f7yyLWnmqPc1cNydHT5BQ,791
-synth_ai/demos/demo_task_apps/math/modal_task_app.py,sha256=eJtNUZvO0foucqQ5-jHOygu3VaffrUEHn9jd1_9ncdM,20112
+synth_ai/demos/demo_task_apps/math/modal_task_app.py,sha256=nKrYZBtehAC3J9WHsmV9V2iiC7vbexEz1rTkOoU0wDI,20180
 synth_ai/environments/__init__.py,sha256=BQW0Nc_BFQq_N-pcqTyJVjW56kSEXu7XZyaSer-U95Q,1032
 synth_ai/environments/environment/__init__.py,sha256=EBol9AKxPTIPXWcbH9Tja-l3yL-N2kB8e5atyf6F66c,31
 synth_ai/environments/environment/core.py,sha256=0jd0CZ88_s_qqA3d1lOgVsnv-ucw_1lJDAIUj1gTSt0,2201
@@ -412,9 +412,9 @@ synth_ai/v0/tracing_v1/events/manage.py,sha256=ZDXXP-ZwLH9LCsmw7Ru9o55d7bl_diPtJ
 synth_ai/v0/tracing_v1/events/scope.py,sha256=BuBkhSpVHUJt8iGT9HJZF82rbb88mQcd2vM2shg-w2I,2550
 synth_ai/v0/tracing_v1/events/store.py,sha256=0342lvAcalyJbVEIzQFaPuMQGgwiFm7M5rE6gr-G0E8,9041
 synth_ai/zyk/__init__.py,sha256=htVLnzTYQ5rxzYpzSYBm7_o6uNKZ3pB_PrqkBrgTRS4,771
-synth_ai-0.2.8.dev11.dist-info/licenses/LICENSE,sha256=ynhjRQUfqA_RdGRATApfFA_fBAy9cno04sLtLUqxVFM,1069
-synth_ai-0.2.8.dev11.dist-info/METADATA,sha256=ibi9UBO3d-6_HewGfYvGBR01F-HxD5G-lYV-on2Gp8s,5153
-synth_ai-0.2.8.dev11.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-synth_ai-0.2.8.dev11.dist-info/entry_points.txt,sha256=Neq-3bT7TAijjgOIR77pKL-WYg6TWBDeO8pp_nL4vGY,91
-synth_ai-0.2.8.dev11.dist-info/top_level.txt,sha256=fBmtZyVHuKaGa29oHBaaUkrUIWTqSpoVMPiVdCDP3k8,9
-synth_ai-0.2.8.dev11.dist-info/RECORD,,
+synth_ai-0.2.8.dev12.dist-info/licenses/LICENSE,sha256=ynhjRQUfqA_RdGRATApfFA_fBAy9cno04sLtLUqxVFM,1069
+synth_ai-0.2.8.dev12.dist-info/METADATA,sha256=quM_ZNMni7xFu7ZSTwA06eAcZbzJjNxbZgrNFDcn22Q,5153
+synth_ai-0.2.8.dev12.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+synth_ai-0.2.8.dev12.dist-info/entry_points.txt,sha256=Neq-3bT7TAijjgOIR77pKL-WYg6TWBDeO8pp_nL4vGY,91
+synth_ai-0.2.8.dev12.dist-info/top_level.txt,sha256=fBmtZyVHuKaGa29oHBaaUkrUIWTqSpoVMPiVdCDP3k8,9
+synth_ai-0.2.8.dev12.dist-info/RECORD,,