multi-forge 1.0.0__tar.gz → 1.0.2__tar.gz

{multi_forge-1.0.0 → multi_forge-1.0.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: multi-forge
-Version: 1.0.0
+Version: 1.0.2
 Summary: Multi-runtime agent toolkit: proxy routing, cost control, session management, policy enforcement, and workflow orchestration
 Project-URL: Homepage, https://github.com/hapa1i/multi-forge
 Project-URL: Repository, https://github.com/hapa1i/multi-forge
@@ -33,7 +33,7 @@ Requires-Dist: expression<6,>=5.6.0
 Requires-Dist: fastapi-sso<1,>=0.19.0
 Requires-Dist: fastapi<0.137,>=0.136.3
 Requires-Dist: httpx>=0.25.0
-Requires-Dist: litellm<=1.99.0,>=1.98.0
+Requires-Dist: litellm==1.102.0
 Requires-Dist: openai>=1.0.0
 Requires-Dist: orjson<4,>=3.11.6
 Requires-Dist: packaging>=21.0

{multi_forge-1.0.0 → multi_forge-1.0.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "multi-forge"
-version = "1.0.0"
+version = "1.0.2"
 description = "Multi-runtime agent toolkit: proxy routing, cost control, session management, policy enforcement, and workflow orchestration"
 readme = "README.md"
 license = "Apache-2.0"
@@ -42,7 +42,7 @@ dependencies = [
     # Proxy dependencies (previously optional, now core). Forge owns the
     # default-Uvicorn LiteLLM startup set because the upstream `proxy` extra
     # caps cryptography below Forge's patched floor. This set is start/health
-    # validated through LiteLLM 1.99.0; run scripts/test-wheel-runtime.sh before
+    # validated through LiteLLM 1.102.0; run scripts/test-wheel-runtime.sh before
     # raising the ceiling. Alternate servers, databases, and cloud integrations
     # remain outside Forge's runtime path.
     "aiohttp>=3.14.3,<4",
@@ -68,10 +68,8 @@ dependencies = [
     "uvloop>=0.21.0,<1; sys_platform != 'win32'",
     "watchfiles>=0.20",
     "websockets>=15.0.1,<16",
-    # Floor = the first packaged-cost-map version for gemini-3.7-flash (see
-    # tests/integration/proxy/test_proxy_local_litellm_e2e.py); packaged
-    # cost-map pricing for that model first ships in the v1.98 line.
-    "litellm>=1.98.0,<=1.99.0",
+    # Packaged Gemini 3.8 Flash metadata and the verified proxy runtime.
+    "litellm==1.102.0",
 ]
 
 [project.scripts]

{multi_forge-1.0.0 → multi_forge-1.0.2}/src/forge/cli/claude.py

@@ -129,8 +129,9 @@ def _build_bare_launch_env(
         env_vars["ANTHROPIC_BASE_URL"] = base_url
         if context_limit is not None:
             env_vars["CLAUDE_CODE_AUTO_COMPACT_WINDOW"] = str(context_limit)
-        apply_proxy_context_model_defaults(env_vars, context_limit)
-        if wire_shape := resolve_proxy_wire_shape(proxy_id=proxy_id, template=template):
+        wire_shape = resolve_proxy_wire_shape(proxy_id=proxy_id, template=template)
+        apply_proxy_context_model_defaults(env_vars, context_limit, wire_shape=wire_shape)
+        if wire_shape:
             env_vars[FORGE_PROXY_WIRE_SHAPE_VAR] = wire_shape
         if template:
             env_vars["ACTIVE_TEMPLATE"] = template

multi_forge-1.0.2/src/forge/config/defaults/backends/litellm.yaml

@@ -0,0 +1,547 @@
+# LiteLLM Local Proxy Configuration
+# For use with personal API keys (Gemini, Anthropic, and/or OpenAI)
+#
+# Supported models:
+#   Gemini (requires GEMINI_API_KEY):
+#     - gemini/gemini-3.1-pro-preview
+#     - gemini/gemini-3.1-pro-preview-customtools (optimized for custom tools + bash)
+#     - gemini/gemini-3-flash-preview
+#     - gemini/gemini-3.8-flash (LiteLLM Flash default)
+#     - gemini/gemini-3.7-flash
+#     - gemini/gemini-3.6-flash (Gemini 3.6 Flash - efficiency release)
+#     - gemini/gemini-3.5-flash
+#     - gemini/gemini-2.5-flash
+#     - gemini/gemini-2.5-pro
+#
+#   Anthropic (requires ANTHROPIC_API_KEY):
+#     - anthropic/claude-haiku-4-5, claude-sonnet-5-5, claude-sonnet-5, claude-opus-5-5, claude-opus-5
+#     - anthropic/claude-fable-5-1, claude-fable-5
+#     - anthropic/claude-opus-4-8, claude-opus-4-6, claude-sonnet-4-6
+#
+#   OpenAI (requires OPENAI_API_KEY):
+#     - openai/gpt-5, gpt-5-codex, gpt-5-mini, gpt-5-nano, gpt-5-pro
+#     - openai/gpt-5.1, gpt-5.1-codex, gpt-5.1-codex-max, gpt-5.1-codex-mini
+#     - openai/gpt-5.2, gpt-5.2-codex, gpt-5.2-pro
+#     - openai/gpt-5.3-codex
+#     - openai/gpt-5.4, gpt-5.4-pro
+#     - openai/gpt-5.5
+#     - openai/gpt-5.6, gpt-5.6-sol, gpt-5.6-terra, gpt-5.6-luna
+#     - openai/gpt-6-astra, gpt-6.1-sol, gpt-6-sol, gpt-6-luna
+#
+# Start with: forge model backend start litellm --port 4000
+
+model_list:
+  # Gemini 3.1 Pro Preview
+  - model_name: gemini/gemini-3.1-pro-preview
+    litellm_params:
+      model: gemini/gemini-3.1-pro-preview
+      api_key: os.environ/GEMINI_API_KEY
+
+  # Gemini 3.1 Pro Preview (Custom Tools) - optimized for agentic workflows with custom tools + bash
+  - model_name: gemini/gemini-3.1-pro-preview-customtools
+    litellm_params:
+      model: gemini/gemini-3.1-pro-preview-customtools
+      api_key: os.environ/GEMINI_API_KEY
+
+  # Gemini 3 Flash Preview
+  - model_name: gemini/gemini-3-flash-preview
+    litellm_params:
+      model: gemini/gemini-3-flash-preview
+      api_key: os.environ/GEMINI_API_KEY
+
+  # Gemini 3.8 Flash - current LiteLLM Flash default
+  - model_name: gemini/gemini-3.8-flash
+    litellm_params:
+      model: gemini/gemini-3.8-flash
+      api_key: os.environ/GEMINI_API_KEY
+
+  # Gemini 3.7 Flash
+  - model_name: gemini/gemini-3.7-flash
+    litellm_params:
+      model: gemini/gemini-3.7-flash
+      api_key: os.environ/GEMINI_API_KEY
+
+  # Gemini 3.6 Flash - efficiency release built on 3.5 Flash
+  - model_name: gemini/gemini-3.6-flash
+    litellm_params:
+      model: gemini/gemini-3.6-flash
+      api_key: os.environ/GEMINI_API_KEY
+
+  # Gemini 3.5 Flash
+  - model_name: gemini/gemini-3.5-flash
+    litellm_params:
+      model: gemini/gemini-3.5-flash
+      api_key: os.environ/GEMINI_API_KEY
+
+  # Gemini 2.5 Flash
+  - model_name: gemini/gemini-2.5-flash
+    litellm_params:
+      model: gemini/gemini-2.5-flash
+      api_key: os.environ/GEMINI_API_KEY
+
+  # Gemini 2.5 Pro
+  - model_name: gemini/gemini-2.5-pro
+    litellm_params:
+      model: gemini/gemini-2.5-pro
+      api_key: os.environ/GEMINI_API_KEY
+
+  # Aliases for convenience (match config.yaml model names)
+  - model_name: gemini-3.1-pro-preview
+    litellm_params:
+      model: gemini/gemini-3.1-pro-preview
+      api_key: os.environ/GEMINI_API_KEY
+
+  - model_name: gemini-3.1-pro-preview-customtools
+    litellm_params:
+      model: gemini/gemini-3.1-pro-preview-customtools
+      api_key: os.environ/GEMINI_API_KEY
+
+  - model_name: gemini-3-flash-preview
+    litellm_params:
+      model: gemini/gemini-3-flash-preview
+      api_key: os.environ/GEMINI_API_KEY
+
+  - model_name: gemini-3.8-flash
+    litellm_params:
+      model: gemini/gemini-3.8-flash
+      api_key: os.environ/GEMINI_API_KEY
+
+  - model_name: gemini-3.7-flash
+    litellm_params:
+      model: gemini/gemini-3.7-flash
+      api_key: os.environ/GEMINI_API_KEY
+
+  - model_name: gemini-3.6-flash
+    litellm_params:
+      model: gemini/gemini-3.6-flash
+      api_key: os.environ/GEMINI_API_KEY
+
+  - model_name: gemini-3.5-flash
+    litellm_params:
+      model: gemini/gemini-3.5-flash
+      api_key: os.environ/GEMINI_API_KEY
+
+  - model_name: gemini-2.5-flash
+    litellm_params:
+      model: gemini/gemini-2.5-flash
+      api_key: os.environ/GEMINI_API_KEY
+
+  - model_name: gemini-2.5-pro
+    litellm_params:
+      model: gemini/gemini-2.5-pro
+      api_key: os.environ/GEMINI_API_KEY
+
+  # =========================================================================
+  # Anthropic Models (requires ANTHROPIC_API_KEY)
+  # =========================================================================
+
+  - model_name: anthropic/claude-haiku-4-5
+    litellm_params:
+      model: anthropic/claude-haiku-4-5
+      api_key: os.environ/ANTHROPIC_API_KEY
+
+  # LiteLLM 1.102 predates Sonnet 5.5; supply offline pricing and effort capabilities.
+  # https://platform.claude.com/docs/en/models/sonnet-5-5/overview (verified 2026-10-01)
+  - model_name: anthropic/claude-sonnet-5-5
+    litellm_params:
+      model: anthropic/claude-sonnet-5-5
+      api_key: os.environ/ANTHROPIC_API_KEY
+    model_info:
+      base_model: claude-sonnet-5-5
+      mode: chat
+      max_input_tokens: 1000000
+      max_output_tokens: 128000
+      max_tokens: 128000
+      input_cost_per_token: 0.000002
+      output_cost_per_token: 0.000010
+      cache_creation_input_token_cost: 0.0000025
+      cache_creation_input_token_cost_above_1hr: 0.000004
+      cache_read_input_token_cost: 0.0000002
+      supports_vision: true
+      supports_function_calling: true
+      supports_tool_choice: true
+      supports_prompt_caching: true
+      prompt_cache_min_tokens: 512
+      supports_reasoning: true
+      supports_adaptive_thinking: true
+      supports_legacy_thinking: false
+      # Keep disabled visible to upstream validation; dropping it would turn thinking on.
+      thinking_always_on: false
+      supports_output_config: true
+      supports_xhigh_reasoning_effort: true
+      supports_max_reasoning_effort: true
+      default_reasoning_effort: high
+      reasoning_effort_levels: [low, medium, high, xhigh, max]
+      supports_response_schema: true
+      supports_native_structured_output: true
+      supports_assistant_prefill: false
+      provider_specific_entry:
+        us: 1.1
+
+  - model_name: anthropic/claude-sonnet-5
+    litellm_params:
+      model: anthropic/claude-sonnet-5
+      api_key: os.environ/ANTHROPIC_API_KEY
+
+  # LiteLLM 1.102 predates Opus 5.5. Keep its pricing and thinking contract available offline.
+  # https://platform.claude.com/docs/en/models/opus-5-5/overview (verified 2026-09-23)
+  - model_name: anthropic/claude-opus-5-5
+    litellm_params:
+      model: anthropic/claude-opus-5-5
+      api_key: os.environ/ANTHROPIC_API_KEY
+    model_info:
+      base_model: claude-opus-5-5
+      mode: chat
+      max_input_tokens: 1000000
+      max_output_tokens: 128000
+      max_tokens: 128000
+      input_cost_per_token: 0.000004
+      output_cost_per_token: 0.000020
+      cache_creation_input_token_cost: 0.000005
+      cache_creation_input_token_cost_above_1hr: 0.000008
+      # Opus 5.5 cache reads intentionally cost $0.20/MTok (0.05x base input).
+      cache_read_input_token_cost: 0.0000002
+      supports_vision: true
+      supports_function_calling: true
+      supports_tool_choice: true
+      supports_prompt_caching: true
+      prompt_cache_min_tokens: 512
+      supports_reasoning: true
+      supports_adaptive_thinking: true
+      supports_legacy_thinking: false
+      thinking_always_on: true
+      supports_output_config: true
+      supports_xhigh_reasoning_effort: true
+      supports_max_reasoning_effort: true
+      default_reasoning_effort: medium
+      reasoning_effort_levels: [low, medium, high, xhigh, max]
+      supports_response_schema: true
+      supports_native_structured_output: true
+      supports_assistant_prefill: false
+      provider_specific_entry:
+        us: 1.1
+        fast: 2.0
+
+  - model_name: anthropic/claude-opus-5
+    litellm_params:
+      model: anthropic/claude-opus-5
+      api_key: os.environ/ANTHROPIC_API_KEY
+
+  - model_name: anthropic/claude-fable-5-1
+    litellm_params:
+      model: anthropic/claude-fable-5-1
+      api_key: os.environ/ANTHROPIC_API_KEY
+
+  - model_name: anthropic/claude-fable-5
+    litellm_params:
+      model: anthropic/claude-fable-5
+      api_key: os.environ/ANTHROPIC_API_KEY
+
+  - model_name: anthropic/claude-opus-4-8
+    litellm_params:
+      model: anthropic/claude-opus-4-8
+      api_key: os.environ/ANTHROPIC_API_KEY
+
+  - model_name: anthropic/claude-opus-4-6
+    litellm_params:
+      model: anthropic/claude-opus-4-6
+      api_key: os.environ/ANTHROPIC_API_KEY
+
+  - model_name: anthropic/claude-sonnet-4-6
+    litellm_params:
+      model: anthropic/claude-sonnet-4-6
+      api_key: os.environ/ANTHROPIC_API_KEY
+
+  # =========================================================================
+  # OpenAI Models (requires OPENAI_API_KEY)
+  # =========================================================================
+
+  # --- GPT-5 family ---
+  - model_name: openai/gpt-5
+    litellm_params:
+      model: openai/gpt-5
+      api_key: os.environ/OPENAI_API_KEY
+
+  - model_name: openai/gpt-5-codex
+    litellm_params:
+      model: openai/gpt-5-codex
+      api_key: os.environ/OPENAI_API_KEY
+
+  - model_name: openai/gpt-5-mini
+    litellm_params:
+      model: openai/gpt-5-mini
+      api_key: os.environ/OPENAI_API_KEY
+
+  - model_name: openai/gpt-5-nano
+    litellm_params:
+      model: openai/gpt-5-nano
+      api_key: os.environ/OPENAI_API_KEY
+
+  - model_name: openai/gpt-5-pro
+    litellm_params:
+      model: openai/gpt-5-pro
+      api_key: os.environ/OPENAI_API_KEY
+
+  # --- GPT-5.1 family ---
+  - model_name: openai/gpt-5.1
+    litellm_params:
+      model: openai/gpt-5.1
+      api_key: os.environ/OPENAI_API_KEY
+
+  - model_name: openai/gpt-5.1-codex
+    litellm_params:
+      model: openai/gpt-5.1-codex
+      api_key: os.environ/OPENAI_API_KEY
+
+  - model_name: openai/gpt-5.1-codex-max
+    litellm_params:
+      model: openai/gpt-5.1-codex-max
+      api_key: os.environ/OPENAI_API_KEY
+
+  - model_name: openai/gpt-5.1-codex-mini
+    litellm_params:
+      model: openai/gpt-5.1-codex-mini
+      api_key: os.environ/OPENAI_API_KEY
+
+  # --- GPT-5.2 family ---
+  - model_name: openai/gpt-5.2
+    litellm_params:
+      model: openai/gpt-5.2
+      api_key: os.environ/OPENAI_API_KEY
+
+  - model_name: openai/gpt-5.2-codex
+    litellm_params:
+      model: openai/gpt-5.2-codex
+      api_key: os.environ/OPENAI_API_KEY
+
+  - model_name: openai/gpt-5.2-pro
+    litellm_params:
+      model: openai/gpt-5.2-pro
+      api_key: os.environ/OPENAI_API_KEY
+
+  # --- GPT-5.3 family ---
+  - model_name: openai/gpt-5.3-codex
+    litellm_params:
+      model: openai/gpt-5.3-codex
+      api_key: os.environ/OPENAI_API_KEY
+
+  # --- GPT-5.4 family ---
+  - model_name: openai/gpt-5.4
+    litellm_params:
+      model: openai/gpt-5.4
+      api_key: os.environ/OPENAI_API_KEY
+
+  - model_name: openai/gpt-5.4-pro
+    litellm_params:
+      model: openai/gpt-5.4-pro
+      api_key: os.environ/OPENAI_API_KEY
+
+  - model_name: openai/gpt-5.4-mini
+    litellm_params:
+      model: openai/gpt-5.4-mini
+      api_key: os.environ/OPENAI_API_KEY
+
+  - model_name: openai/gpt-5.4-nano
+    litellm_params:
+      model: openai/gpt-5.4-nano
+      api_key: os.environ/OPENAI_API_KEY
+
+  # --- GPT-5.5 ---
+  - model_name: openai/gpt-5.5
+    litellm_params:
+      model: openai/gpt-5.5
+      api_key: os.environ/OPENAI_API_KEY
+
+  # --- GPT-5.6 family ---
+  - model_name: openai/gpt-5.6
+    litellm_params:
+      model: openai/gpt-5.6
+      api_key: os.environ/OPENAI_API_KEY
+
+  - model_name: openai/gpt-5.6-sol
+    litellm_params:
+      model: openai/gpt-5.6-sol
+      api_key: os.environ/OPENAI_API_KEY
+
+  - model_name: openai/gpt-5.6-terra
+    litellm_params:
+      model: openai/gpt-5.6-terra
+      api_key: os.environ/OPENAI_API_KEY
+
+  - model_name: openai/gpt-5.6-luna
+    litellm_params:
+      model: openai/gpt-5.6-luna
+      api_key: os.environ/OPENAI_API_KEY
+
+  - model_name: openai/gpt-6-astra
+    litellm_params:
+      model: openai/gpt-6-astra
+      api_key: os.environ/OPENAI_API_KEY
+
+  # LiteLLM 1.102 predates GPT-6.1 Sol; cached reads are 5% of input, including the >272k premium.
+  # https://developers.openai.com/api/docs/models/gpt-6.1-sol (verified 2026-10-01)
+  - model_name: openai/gpt-6.1-sol
+    litellm_params:
+      model: openai/gpt-6.1-sol
+      api_key: os.environ/OPENAI_API_KEY
+      use_responses_api: true
+    model_info:
+      base_model: gpt-6.1-sol
+      mode: responses
+      max_input_tokens: 922000
+      max_output_tokens: 128000
+      max_tokens: 128000
+      input_cost_per_token: 0.000002
+      output_cost_per_token: 0.000010
+      cache_read_input_token_cost: 0.0000001
+      cache_creation_input_token_cost: 0.0000025
+      input_cost_per_token_above_272k_tokens: 0.000004
+      output_cost_per_token_above_272k_tokens: 0.000015
+      cache_read_input_token_cost_above_272k_tokens: 0.0000002
+      cache_creation_input_token_cost_above_272k_tokens: 0.000005
+      input_cost_per_token_flex: 0.000001
+      output_cost_per_token_flex: 0.000005
+      cache_read_input_token_cost_flex: 0.00000005
+      cache_creation_input_token_cost_flex: 0.00000125
+      input_cost_per_token_above_272k_tokens_flex: 0.000002
+      output_cost_per_token_above_272k_tokens_flex: 0.0000075
+      cache_read_input_token_cost_above_272k_tokens_flex: 0.0000001
+      cache_creation_input_token_cost_above_272k_tokens_flex: 0.0000025
+      input_cost_per_token_priority: 0.000004
+      output_cost_per_token_priority: 0.000020
+      cache_read_input_token_cost_priority: 0.0000002
+      cache_creation_input_token_cost_priority: 0.000005
+      input_cost_per_token_above_272k_tokens_priority: 0.000008
+      output_cost_per_token_above_272k_tokens_priority: 0.000030
+      cache_read_input_token_cost_above_272k_tokens_priority: 0.0000004
+      cache_creation_input_token_cost_above_272k_tokens_priority: 0.000010
+      supports_vision: true
+      supports_function_calling: true
+      supports_parallel_function_calling: true
+      supports_tool_choice: true
+      supports_prompt_caching: true
+      supports_reasoning: true
+      supports_none_reasoning_effort: false
+      supports_xhigh_reasoning_effort: true
+      supports_max_reasoning_effort: true
+      default_reasoning_effort: medium
+      reasoning_effort_levels: [low, medium, high, xhigh, max]
+      supports_response_schema: true
+
+  # LiteLLM 1.102 predates Sol and Luna. Standard, Flex and Fast rates include the >272k-token premium.
+  # Flex costs half of standard; Fast (also called priority) costs twice the applicable rate.
+  # Remove these overrides when bundled metadata covers them; test_litellm_model_metadata pins that boundary.
+  # https://developers.openai.com/api/docs/models/gpt-6-sol (verified 2026-09-23)
+  - model_name: openai/gpt-6-sol
+    litellm_params:
+      model: openai/gpt-6-sol
+      api_key: os.environ/OPENAI_API_KEY
+      use_responses_api: true
+    model_info:
+      base_model: gpt-6-sol
+      mode: responses
+      max_input_tokens: 922000
+      max_output_tokens: 128000
+      max_tokens: 128000
+      input_cost_per_token: 0.000002
+      output_cost_per_token: 0.000010
+      cache_read_input_token_cost: 0.0000002
+      cache_creation_input_token_cost: 0.0000025
+      input_cost_per_token_above_272k_tokens: 0.000004
+      output_cost_per_token_above_272k_tokens: 0.000015
+      cache_read_input_token_cost_above_272k_tokens: 0.0000004
+      cache_creation_input_token_cost_above_272k_tokens: 0.000005
+      input_cost_per_token_flex: 0.000001
+      output_cost_per_token_flex: 0.000005
+      cache_read_input_token_cost_flex: 0.0000001
+      cache_creation_input_token_cost_flex: 0.00000125
+      input_cost_per_token_above_272k_tokens_flex: 0.000002
+      output_cost_per_token_above_272k_tokens_flex: 0.0000075
+      cache_read_input_token_cost_above_272k_tokens_flex: 0.0000002
+      cache_creation_input_token_cost_above_272k_tokens_flex: 0.0000025
+      input_cost_per_token_priority: 0.000004
+      output_cost_per_token_priority: 0.000020
+      cache_read_input_token_cost_priority: 0.0000004
+      cache_creation_input_token_cost_priority: 0.000005
+      input_cost_per_token_above_272k_tokens_priority: 0.000008
+      output_cost_per_token_above_272k_tokens_priority: 0.000030
+      cache_read_input_token_cost_above_272k_tokens_priority: 0.0000008
+      cache_creation_input_token_cost_above_272k_tokens_priority: 0.000010
+      supports_vision: true
+      supports_function_calling: true
+      supports_parallel_function_calling: true
+      supports_tool_choice: true
+      supports_prompt_caching: true
+      supports_reasoning: true
+      supports_none_reasoning_effort: true
+      supports_xhigh_reasoning_effort: true
+      supports_max_reasoning_effort: true
+      default_reasoning_effort: medium
+      reasoning_effort_levels: [none, low, medium, high, xhigh, max]
+      supports_response_schema: true
+
+  # https://developers.openai.com/api/docs/models/gpt-6-luna (verified 2026-09-23)
+  - model_name: openai/gpt-6-luna
+    litellm_params:
+      model: openai/gpt-6-luna
+      api_key: os.environ/OPENAI_API_KEY
+      use_responses_api: true
+    model_info:
+      base_model: gpt-6-luna
+      mode: responses
+      max_input_tokens: 922000
+      max_output_tokens: 128000
+      max_tokens: 128000
+      input_cost_per_token: 0.0000001
+      output_cost_per_token: 0.0000005
+      cache_read_input_token_cost: 0.00000001
+      cache_creation_input_token_cost: 0.000000125
+      input_cost_per_token_above_272k_tokens: 0.0000002
+      output_cost_per_token_above_272k_tokens: 0.00000075
+      cache_read_input_token_cost_above_272k_tokens: 0.00000002
+      cache_creation_input_token_cost_above_272k_tokens: 0.00000025
+      input_cost_per_token_flex: 0.00000005
+      output_cost_per_token_flex: 0.00000025
+      cache_read_input_token_cost_flex: 0.000000005
+      cache_creation_input_token_cost_flex: 0.0000000625
+      input_cost_per_token_above_272k_tokens_flex: 0.0000001
+      output_cost_per_token_above_272k_tokens_flex: 0.000000375
+      cache_read_input_token_cost_above_272k_tokens_flex: 0.00000001
+      cache_creation_input_token_cost_above_272k_tokens_flex: 0.000000125
+      input_cost_per_token_priority: 0.0000002
+      output_cost_per_token_priority: 0.000001
+      cache_read_input_token_cost_priority: 0.00000002
+      cache_creation_input_token_cost_priority: 0.00000025
+      input_cost_per_token_above_272k_tokens_priority: 0.0000004
+      output_cost_per_token_above_272k_tokens_priority: 0.0000015
+      cache_read_input_token_cost_above_272k_tokens_priority: 0.00000004
+      cache_creation_input_token_cost_above_272k_tokens_priority: 0.0000005
+      supports_vision: true
+      supports_function_calling: true
+      supports_parallel_function_calling: true
+      supports_tool_choice: true
+      supports_prompt_caching: true
+      supports_reasoning: true
+      supports_none_reasoning_effort: true
+      supports_xhigh_reasoning_effort: true
+      supports_max_reasoning_effort: true
+      default_reasoning_effort: medium
+      reasoning_effort_levels: [none, low, medium, high, xhigh, max]
+      supports_response_schema: true
+
+litellm_settings:
+  # Drop unmapped params instead of erroring
+  drop_params: true
+
+  set_verbose: false
+
+  request_timeout: 600
+
+# General settings (optional)
+# general_settings:
+#   # Master key for API authentication (optional)
+#   # Set via: export LITELLM_MASTER_KEY="sk-your-key"
+#   master_key: os.environ/LITELLM_MASTER_KEY
+#
+#   # Database for tracking (optional)
+#   # database_url: "postgresql://..."

{multi_forge-1.0.0 → multi_forge-1.0.2}/src/forge/config/defaults/templates/anthropic-passthrough.yaml

@@ -27,5 +27,5 @@ proxy:
   litellm:
     tiers:
       haiku: claude-haiku-4-5
-      sonnet: claude-sonnet-5
-      opus: claude-opus-5
+      sonnet: claude-sonnet-5-5
+      opus: claude-opus-5-5

{multi_forge-1.0.0 → multi_forge-1.0.2}/src/forge/config/defaults/templates/codex-responses-local.yaml

@@ -27,7 +27,15 @@ proxy:
       sonnet: openai/gpt-6-astra
       opus: openai/gpt-6-astra
     model_alternatives:
+      haiku:
+        gpt-6-luna: openai/gpt-6-luna
       sonnet:
+        gpt-6.1-sol: openai/gpt-6.1-sol
+        gpt-6-sol: openai/gpt-6-sol
+        gpt-6-luna: openai/gpt-6-luna
         gpt-5.6-sol: openai/gpt-5.6-sol
       opus:
+        gpt-6.1-sol: openai/gpt-6.1-sol
+        gpt-6-sol: openai/gpt-6-sol
+        gpt-6-luna: openai/gpt-6-luna
         gpt-5.6-sol: openai/gpt-5.6-sol

{multi_forge-1.0.0 → multi_forge-1.0.2}/src/forge/config/defaults/templates/litellm-anthropic-local.yaml

@@ -6,9 +6,9 @@
 #
 # Backend (LiteLLM on port 4000) starts automatically.
 #
-# Defaults: sonnet tier -> Sonnet 5, opus tier -> Opus 5. Alternatives are
-# claude-fable-5-1, claude-fable-5, claude-opus-4-8, claude-opus-4-6,
-# and claude-sonnet-4-6.
+# Defaults: sonnet tier -> Sonnet 5.5, opus tier -> Opus 5.5. Alternatives are
+# claude-opus-5, claude-fable-5-1, claude-fable-5, claude-opus-4-8, claude-opus-4-6,
+# claude-sonnet-5, and claude-sonnet-4-6.
 #
 # After adding ANTHROPIC_API_KEY, stop the backend so Forge restarts it with the new environment:
 # forge model backend stop litellm-4000
@@ -22,15 +22,17 @@ proxy:
   litellm:
     tiers:
       haiku: anthropic/claude-haiku-4-5
-      sonnet: anthropic/claude-sonnet-5
-      opus: anthropic/claude-opus-5
+      sonnet: anthropic/claude-sonnet-5-5
+      opus: anthropic/claude-opus-5-5
     model_alternatives:
       opus:
+        claude-opus-5: anthropic/claude-opus-5
         claude-fable-5-1: anthropic/claude-fable-5-1
         claude-fable-5: anthropic/claude-fable-5
         claude-opus-4-8: anthropic/claude-opus-4-8
         claude-opus-4-6: anthropic/claude-opus-4-6
       sonnet:
+        claude-sonnet-5: anthropic/claude-sonnet-5
         claude-sonnet-4-6: anthropic/claude-sonnet-4-6
     prompt_caching: passthrough
     error_hints: true