alysis-code 0.13.5__tar.gz → 0.13.6.1__tar.gz

{alysis_code-0.13.5 → alysis_code-0.13.6.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: alysis-code
-Version: 0.13.5
+Version: 0.13.6.1
 Summary: Local CLI coding agent powered by any API.
 Project-URL: Changelog, https://github.com/AlysisAi/alysis-code/blob/main/docs/CHANGELOG.md
 Project-URL: Documentation, https://github.com/AlysisAi/alysis-code/tree/main/docs

{alysis_code-0.13.5 → alysis_code-0.13.6.1}/docs/CHANGELOG.md

@@ -408,6 +408,34 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
   Workspace-target extraction also keeps extensionless dotfiles such as `.npmrc` available to
   verification selection.
 
+## [0.13.6.1] - 2026-09-08
+
+### Fixed
+
+- **V4.1 Flash beta on Alysis Code Pro.** The Pro model picker now includes
+  `deepseek-v4.1-flash-expires-on-0910` and `deepseek-v4-flash-vision-exp`
+  with descriptions. Newly discovered models are captioned "available on your
+  Alysis Code Pro plan" instead of the retired MiMo "trial" wording.
+
+## [0.13.6] - 2026-09-08
+
+### Added
+
+- **DeepSeek V4.1 Flash internal beta.** `deepseek-v4.1-flash-expires-on-0910`
+  (announced 2026-09-08) is now on the DeepSeek preset menu. It is a
+  new-architecture model with native multimodal input, served on the unchanged
+  `https://api.deepseek.com` base URL under a temporary id that stops resolving
+  on 2026-09-10; DeepSeek bills it at the `deepseek-v4-flash` rate and caps
+  accounts at 20 concurrent requests. Registry metadata assumes the V4 Flash
+  shape (1M context, 384K output, vision, thinking `low|high|max`) until
+  DeepSeek publishes the GA id and limits. `deepseek-v4-pro` stays the default.
+
+### Changed
+
+- **DeepSeek footer labels follow vendor spelling.** Names now appear as
+  `DeepSeek-V4-Pro` and `DeepSeek-V4.1-Flash`; dated beta ids omit their
+  `-expires-on-<date>` suffix from the displayed label.
+
 ## [0.13.5] - 2026-09-07
 
 ### Added

{alysis_code-0.13.5 → alysis_code-0.13.6.1}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "alysis-code"
-version = "0.13.5"
+version = "0.13.6.1"
 description = "Local CLI coding agent powered by any API."
 readme = "README.md"
 requires-python = ">=3.11"

{alysis_code-0.13.5 → alysis_code-0.13.6.1}/src/alysis_code/__init__.py

@@ -1,3 +1,3 @@
 __all__ = ["__version__"]
 
-__version__ = "0.13.5"
+__version__ = "0.13.6.1"

{alysis_code-0.13.5 → alysis_code-0.13.6.1}/src/alysis_code/cli_impl/config_menu.py

@@ -3104,8 +3104,9 @@ def _preset_model_option_rows(
 
     NVIDIA's API catalog contains both NVIDIA and third-party models, so its live
     ``/v1/models`` inventory is merged with a small curated recommendation set.
-    The hosted MiMo trial similarly appends its live proxy allowlist. Discovery is
-    best-effort and offline-safe: on failure the static rows still render.
+    The hosted Alysis Code Pro gateway similarly appends its live ``/v1/models``
+    allowlist. Discovery is best-effort and offline-safe: on failure the static
+    rows still render.
     """
     rows = list(model_options_for_preset(preset))
 
@@ -3155,7 +3156,10 @@ def _preset_model_option_rows(
         if any(_same_model(model_id, existing) for existing in known):
             continue
         known.append(model_id)
-        rows.append((model_id, model_id, "available on your Alysis Code trial"))
+        # A model the gateway serves but the client predates: the preset may
+        # still describe it; otherwise say where it comes from.
+        description = str(preset.suggested_model_descriptions.get(model_id) or "").strip()
+        rows.append((model_id, model_id, description or "available on your Alysis Code Pro plan"))
     return rows
 
 

{alysis_code-0.13.5 → alysis_code-0.13.6.1}/src/alysis_code/cli_impl/tui/content.py

@@ -40,18 +40,27 @@ _ACRONYMS = {
 # Flash", "Claude Opus 5", "Kimi K3"), which matches those vendors' naming.
 _HYPHENATED_VERSION_FAMILIES = frozenset({"GPT", "GLM"})
 
+# Families whose vendors hyphenate every word of the name ("DeepSeek-V4-Flash",
+# "DeepSeek-V4.1-Flash", "DeepSeek-V4-Flash-Vision-Exp").
+_FULLY_HYPHENATED_FAMILIES = frozenset({"DeepSeek"})
+
+# DeepSeek serves betas under dated ids ("deepseek-v4.1-flash-expires-on-0910",
+# "v3.2_speciale_expires_on_20251215"). The expiry is routing detail, not part
+# of the model's name.
+_EXPIRY_SUFFIX = re.compile(r"[-_]expires[-_]on[-_]\d+$", re.IGNORECASE)
+
 
 def pretty_model_label(model: str | None) -> str:
     """Turn a raw model id into a friendly footer label.
 
-    ``deepseek-chat`` -> ``DeepSeek Chat``; ``gpt-6-astra`` -> ``GPT-6 Astra``;
-    ``gpt-4o`` -> ``GPT-4o``. Falls back to the raw id when there is nothing
-    sensible to do.
+    ``deepseek-v4.1-flash`` -> ``DeepSeek-V4.1-Flash``; ``gpt-6-astra`` ->
+    ``GPT-6 Astra``; ``gpt-4o`` -> ``GPT-4o``. Falls back to the raw id when
+    there is nothing sensible to do.
     """
     raw = (model or "").strip()
     if not raw:
         return "model"
-    name = raw.rsplit("/", 1)[-1]
+    name = _EXPIRY_SUFFIX.sub("", raw.rsplit("/", 1)[-1])
     tokens = [t for t in re.split(r"[-_\s]+", name) if t]
     out: list[str] = []
     for token in tokens:
@@ -71,7 +80,11 @@ def pretty_model_label(model: str | None) -> str:
             out[-1] = f"{out[-1]}.{word}"
             continue
         out.append(word)
-    return " ".join(out) or raw
+    if not out:
+        return raw
+    if out[0] in _FULLY_HYPHENATED_FAMILIES:
+        return "-".join(out)
+    return " ".join(out)
 
 
 def heading_fragments() -> FormattedText:

{alysis_code-0.13.5 → alysis_code-0.13.6.1}/src/alysis_code/model_registry.py

@@ -92,6 +92,10 @@ _CANONICAL_MODEL_SOURCES: dict[str, tuple[str, ...]] = {
     "deepseek-v4-pro": ("https://api-docs.deepseek.com/quick_start/pricing/",),
     "deepseek-v4-flash": ("https://api-docs.deepseek.com/quick_start/pricing/",),
     "deepseek-v4-flash-vision-exp": ("https://api-docs.deepseek.com/quick_start/pricing/",),
+    "deepseek-v4.1-flash-expires-on-0910": (
+        "https://news.ycombinator.com/item?id=49607094",
+        "https://api-docs.deepseek.com/quick_start/pricing/",
+    ),
     "glm-5.3": (
         "https://docs.z.ai/guides/llm/glm-5.3",
         "https://docs.bigmodel.cn/cn/guide/models/text/glm-5.3",
@@ -151,6 +155,16 @@ _CANONICAL_MODEL_METADATA: dict[str, dict[str, Any]] = {
         "supports_vision": True,
         "supports_reasoning": True,
     },
+    # DeepSeek-V4.1-Flash internal beta (2026-09-08): new architecture with
+    # native multimodal input, served on the same base_url under a temporary
+    # id that expires 2026-09-10. DeepSeek publishes no separate limits, so the
+    # V4 Flash capacity is assumed until the GA id and docs land.
+    "deepseek-v4.1-flash-expires-on-0910": {
+        "context_window_tokens": 1_000_000,
+        "max_output_tokens": 384_000,
+        "supports_vision": True,
+        "supports_reasoning": True,
+    },
     # GLM-5.3 (2026-08-19) and GLM-5.3-Flash (2026-08-26): 1M context, 128K
     # output, thinking cannot be disabled. Flash is natively multimodal.
     "glm-5.3": {
@@ -188,6 +202,7 @@ _PROVIDER_CANONICAL_MODEL_IDS: dict[str, dict[str, str]] = {
         "deepseek-v4-pro": "deepseek-v4-pro",
         "deepseek-v4-flash": "deepseek-v4-flash",
         "deepseek-v4-flash-vision-exp": "deepseek-v4-flash-vision-exp",
+        "deepseek-v4.1-flash-expires-on-0910": "deepseek-v4.1-flash-expires-on-0910",
     },
     "zhipu": {
         "glm-5.3": "glm-5.3",
@@ -314,6 +329,7 @@ _OFFICIAL_PROVIDER_MODEL_SOURCES: dict[str, dict[str, tuple[str, ...]]] = {
             "deepseek-v4-pro",
             "deepseek-v4-flash",
             "deepseek-v4-flash-vision-exp",
+            "deepseek-v4.1-flash-expires-on-0910",
         )
     },
     "openrouter": {
@@ -762,6 +778,13 @@ _OFFICIAL_PROVIDER_MODEL_METADATA: dict[str, dict[str, dict[str, Any]]] = {
             "cache_read_input_cost_per_token": 0.000000014,
             "reasoning_output_cost_per_token": 0.00000132,
         },
+        # Beta is billed at the deepseek-v4-flash rate (announcement, 2026-09-08).
+        "deepseek-v4.1-flash-expires-on-0910": {
+            "input_cost_per_token": 0.00000044,
+            "output_cost_per_token": 0.00000132,
+            "cache_read_input_cost_per_token": 0.000000014,
+            "reasoning_output_cost_per_token": 0.00000132,
+        },
     },
     "openrouter": {
         # Prices are OpenRouter's listed gateway rates (openrouter.ai/api/v1/

{alysis_code-0.13.5 → alysis_code-0.13.6.1}/src/alysis_code/profile_presets.py

@@ -750,6 +750,7 @@ PROFILE_PRESETS: tuple[ProfilePreset, ...] = (
             "deepseek-v4-pro",
             "deepseek-v4-flash",
             "deepseek-v4-flash-vision-exp",
+            "deepseek-v4.1-flash-expires-on-0910",
         ),
         suggested_model_descriptions={
             "deepseek-v4-pro": "default - flagship coding model, 1M context",
@@ -757,6 +758,10 @@ PROFILE_PRESETS: tuple[ProfilePreset, ...] = (
             "deepseek-v4-flash-vision-exp": (
                 "vision preview - image understanding and tools, 1M context"
             ),
+            "deepseek-v4.1-flash-expires-on-0910": (
+                "V4.1 Flash beta - new architecture, native multimodal, priced as "
+                "v4-flash; temporary id expires 2026-09-10"
+            ),
         },
         model_aliases={
             # deepseek-chat / deepseek-reasoner were discontinued 2026-07-24 and
@@ -770,7 +775,9 @@ PROFILE_PRESETS: tuple[ProfilePreset, ...] = (
             "Retired aliases deepseek-chat and deepseek-reasoner no longer resolve; "
             "use the V4 model IDs. Pricing is peak/off-peak (off-peak is 50% of the "
             "listed rate). The vision model is experimental and may change without "
-            "a stable-release deprecation window."
+            "a stable-release deprecation window. deepseek-v4.1-flash-expires-on-0910 "
+            "is an internal-beta id that stops resolving on 2026-09-10 and is capped "
+            "at 20 concurrent requests per account."
         ),
     ),
     ProfilePreset(
@@ -1691,10 +1698,22 @@ PROFILE_PRESETS: tuple[ProfilePreset, ...] = (
         # The models the subscription offers. Live availability is discovered
         # from the gateway's /v1/models at runtime; this static list is the
         # offline fallback and the menu shown before a model is chosen.
-        suggested_models=("deepseek-v4-flash", "deepseek-v4-pro"),
+        suggested_models=(
+            "deepseek-v4-flash",
+            "deepseek-v4-pro",
+            "deepseek-v4-flash-vision-exp",
+            "deepseek-v4.1-flash-expires-on-0910",
+        ),
         suggested_model_descriptions={
             "deepseek-v4-flash": "default - fast high-volume coding (1M context, free daily allowance)",
             "deepseek-v4-pro": "flagship - deeper reasoning (1M context, requires Alysis Code Pro)",
+            "deepseek-v4-flash-vision-exp": (
+                "vision preview - image understanding at the v4-flash rate (experimental)"
+            ),
+            "deepseek-v4.1-flash-expires-on-0910": (
+                "V4.1 Flash beta - new architecture, native multimodal, v4-flash rate; "
+                "DeepSeek retires this id on 2026-09-10"
+            ),
         },
         validation_model="deepseek-v4-flash",
         # Migrate ids from the retired Xiaomi MiMo trial to the Pro default so