release: v0.2.9 (#447)

* feat(llm): add GLM-5.3-Flash, Qwen3.8-Flash, and Tencent HY4 preview

GLM-5.3-Flash on Zhipu, Zhipu Coding Plan and OpenRouter; Qwen3.8-Flash on
DashScope, DashScope Coding Plan and OpenRouter; HY4 preview on OpenRouter.
All three get explicit 1M-class context-window entries so they are not
caught by the glm-5 family fallback (203K) or the 200K default.

* chore: update wechat_group image asset

* chore: update version to v0.2.9

uv.lock moves to deepagents 0.7.11 with its raised dependency floors
(langchain 1.3.18, langchain-core 1.6.1, langchain-anthropic 1.7.0,
langchain-google-genai 4.3.7, langsmith 0.11.2); langgraph stays 1.2.11.

* docs: use umbrella provider names in v0.2.9 changelog

EvoScientist/llm/context_window.py

@@ -17,8 +17,9 @@ _KNOWN_MODEL_CONTEXT_WINDOWS: dict[str, int] = {
     # Qwen 3.6 open-source variants — exceptions to the ``qwen3.6`` family.
     "qwen3.6-27b": 262_000,
     "qwen3.6-35b-a3b": 262_000,
-    # Qwen 3.8 flagship — Max (1M).
+    # Qwen 3.8 closed-source tiers — Max flagship and Flash (1M).
     "qwen3.8-max": 1_000_000,
+    "qwen3.8-flash": 1_000_000,
     # Qwen 3.7 closed-source tiers — Max flagship and Plus (1M).
     "qwen3.7-max": 1_000_000,
     "qwen3.7-plus": 1_000_000,
@@ -35,10 +36,12 @@ _KNOWN_MODEL_CONTEXT_WINDOWS: dict[str, int] = {
     # Zhipu GLM-5.3/5.2 — 1M context, exceptions to the ``glm-5`` family (203K).
     # Matches OpenRouter ``z-ai/glm-5.x`` via split('/')[-1].
     "glm-5.3": 1_000_000,
+    "glm-5.3-flash": 1_000_000,
     "glm-5.2": 1_000_000,
     # Volcengine Coding Plan's OpenAI-compatible alias for GLM-5.2.
     "glm-5-2": 1_000_000,
-    # Tencent Hunyuan HY3 — 262K context (OpenRouter ``tencent/hy3``).
+    # Tencent Hunyuan — HY4 preview 1M, HY3 262K (OpenRouter ``tencent/hy*``).
+    "hy4-preview": 1_048_576,
     "hy3": 262_000,
 }
 

EvoScientist/llm/registry.py

@@ -185,6 +185,7 @@ _MODEL_ENTRIES: list[tuple[str, str, str]] = [
     ("kimi-k3", "moonshotai/kimi-k3", "openrouter"),
     ("kimi-k2.6", "moonshotai/kimi-k2.6", "openrouter"),
     ("glm-5.3", "z-ai/glm-5.3", "openrouter"),
+    ("glm-5.3-flash", "z-ai/glm-5.3-flash", "openrouter"),
     ("glm-5.2", "z-ai/glm-5.2", "openrouter"),
     ("glm-5v-turbo", "z-ai/glm-5v-turbo", "openrouter"),
     ("minimax-m3", "minimax/minimax-m3", "openrouter"),
@@ -193,8 +194,10 @@ _MODEL_ENTRIES: list[tuple[str, str, str]] = [
     ("grok-build-0.1", "x-ai/grok-build-0.1", "openrouter"),
     ("grok-4.6", "x-ai/grok-4.6", "openrouter"),
     ("grok-4.5", "x-ai/grok-4.5", "openrouter"),
+    ("hy4-preview", "tencent/hy4-preview", "openrouter"),
     ("hy3", "tencent/hy3", "openrouter"),
     ("qwen3.8-max", "qwen/qwen3.8-max", "openrouter"),
+    ("qwen3.8-flash", "qwen/qwen3.8-flash", "openrouter"),
     ("qwen3.7-max", "qwen/qwen3.7-max", "openrouter"),
     ("qwen3.7-plus", "qwen/qwen3.7-plus", "openrouter"),
     ("qwen3.6-flash", "qwen/qwen3.6-flash", "openrouter"),
@@ -208,6 +211,7 @@ _MODEL_ENTRIES: list[tuple[str, str, str]] = [
     ("kimi-k2.5", "kimi-k2-5", "volcengine-code"),
     # Zhipu CodePlan (智谱代码计划 — coding-only endpoint)
     ("glm-5.3", "glm-5.3", "zhipu-code"),
+    ("glm-5.3-flash", "glm-5.3-flash", "zhipu-code"),
     ("glm-5.2", "glm-5.2", "zhipu-code"),
     ("glm-5.1", "glm-5.1", "zhipu-code"),
     ("glm-5", "glm-5", "zhipu-code"),
@@ -216,6 +220,7 @@ _MODEL_ENTRIES: list[tuple[str, str, str]] = [
     ("glm-4.7", "glm-4.7", "zhipu-code"),
     # Zhipu (智谱 — general endpoint, default for simple lookups)
     ("glm-5.3", "glm-5.3", "zhipu"),
+    ("glm-5.3-flash", "glm-5.3-flash", "zhipu"),
     ("glm-5.2", "glm-5.2", "zhipu"),
     ("glm-5.1", "glm-5.1", "zhipu"),
     ("glm-5", "glm-5", "zhipu"),
@@ -232,6 +237,7 @@ _MODEL_ENTRIES: list[tuple[str, str, str]] = [
     ("doubao-1.5-thinking-pro", "doubao-1.5-thinking-pro", "volcengine"),
     # DashScope Coding Plan (阿里云代码计划 — subscription sk-sp-* endpoint)
     ("qwen3.8-max", "qwen3.8-max", "dashscope-code"),
+    ("qwen3.8-flash", "qwen3.8-flash", "dashscope-code"),
     ("qwen3.7-max", "qwen3.7-max", "dashscope-code"),
     ("qwen3.7-plus", "qwen3.7-plus", "dashscope-code"),
     ("qwen3.6-max", "qwen3.6-max-preview", "dashscope-code"),
@@ -243,6 +249,7 @@ _MODEL_ENTRIES: list[tuple[str, str, str]] = [
     ("qwen3.5-plus", "qwen3.5-plus", "dashscope-code"),
     # DashScope (阿里云 — Qwen models, default for simple lookups)
     ("qwen3.8-max", "qwen3.8-max", "dashscope"),
+    ("qwen3.8-flash", "qwen3.8-flash", "dashscope"),
     ("qwen3.7-max", "qwen3.7-max", "dashscope"),
     ("qwen3.7-plus", "qwen3.7-plus", "dashscope"),
     ("qwen3.6-max", "qwen3.6-max-preview", "dashscope"),