hermes-agent/hermes_cli
0xbyt4 4ac731c841 fix(model-normalize): pass DeepSeek V-series IDs through instead of folding to deepseek-chat
`_normalize_for_deepseek` was mapping every non-reasoner input into
`deepseek-chat` on the assumption that DeepSeek's API accepts only two
model IDs. That assumption no longer holds — `deepseek-v4-pro` and
`deepseek-v4-flash` are first-class IDs accepted by the direct API,
and on aggregators `deepseek-chat` routes explicitly to V3 (DeepInfra
backend returns `deepseek-chat-v3`). So a user picking V4 Pro through
the model picker was being silently downgraded to V3.

Verified 2026-04-24 against Nous portal's OpenAI-compat surface:
  - `deepseek/deepseek-v4-flash` → provider: DeepSeek,
    model: deepseek-v4-flash-20260423
  - `deepseek/deepseek-chat`     → provider: DeepInfra,
    model: deepseek/deepseek-chat-v3

Fix:
- Add `deepseek-v4-pro` and `deepseek-v4-flash` to
  `_DEEPSEEK_CANONICAL_MODELS` so exact matches pass through.
- Add `_DEEPSEEK_V_SERIES_RE` (`^deepseek-v\d+(...)?$`) so future
  V-series IDs (`deepseek-v5-*`, dated variants) keep passing through
  without another code change.
- Update docstring + module header to reflect the new rule.

Tests:
- New `TestDeepseekVSeriesPassThrough` — 8 parametrized cases covering
  bare, vendor-prefixed, case-variant, dated, and future V-series IDs
  plus end-to-end `normalize_model_for_provider(..., "deepseek")`.
- New `TestDeepseekCanonicalAndReasonerMapping` — regression coverage
  for canonical pass-through, reasoner-keyword folding, and
  fall-back-to-chat behaviour.
- 77/77 pass.

Reported on Discord (Ufonik, Don Piedro): `/model > Deepseek >
deepseek-v4-pro` surfaced
`Normalized 'deepseek-v4-pro' to 'deepseek-chat'`. Picker listing
showed the v4 names, so validation also rejected the post-normalize
`deepseek-chat` as "not in provider listing" — the contradiction
users saw. Normalizer now respects the picker's choice.
2026-04-24 05:24:54 -07:00
..
__init__.py chore: release v0.11.0 (2026.4.23) (#14791) 2026-04-23 15:31:59 -07:00
auth_commands.py Add native Spotify tools with PKCE auth 2026-04-24 05:20:38 -07:00
auth.py Add native Spotify tools with PKCE auth 2026-04-24 05:20:38 -07:00
backup.py
banner.py feat(banner): hyperlink startup banner title to latest GitHub release (#14945) 2026-04-23 23:28:34 -07:00
callbacks.py
claw.py Normalize claw workspace paths for Windows 2026-04-22 18:15:27 -07:00
cli_output.py
clipboard.py
codex_models.py feat(codex): add gpt-5.5 and wire live model discovery into picker (#14720) 2026-04-23 13:32:43 -07:00
colors.py
commands.py refactor(commands): drop /provider, /plan handler, and clean up slash registry (#15047) 2026-04-24 03:10:52 -07:00
completion.py
config.py feat: read prompt caching cache_ttl from config 2026-04-24 03:21:29 -07:00
copilot_auth.py fix(copilot): exchange raw GitHub token for Copilot API JWT 2026-04-24 05:09:08 -07:00
cron.py feat(cron): per-job workdir for project-aware cron runs (#15110) 2026-04-24 05:07:01 -07:00
curses_ui.py
debug.py style(debug): add missing blank line between LogSnapshot and helpers 2026-04-22 16:34:05 -05:00
default_soul.py
dingtalk_auth.py
doctor.py fix(cli): validate user-defined providers consistently 2026-04-24 04:48:56 -07:00
dump.py
env_loader.py fix(cli): ensure project .env is sanitized before loading 2026-04-22 05:51:44 -07:00
gateway.py fix(gateway): drain-aware hermes update + faster still-working pings (#14736) 2026-04-23 14:01:57 -07:00
hooks.py
logs.py
main.py Add native Spotify tools with PKCE auth 2026-04-24 05:20:38 -07:00
mcp_config.py
memory_setup.py
model_normalize.py fix(model-normalize): pass DeepSeek V-series IDs through instead of folding to deepseek-chat 2026-04-24 05:24:54 -07:00
model_switch.py fix(model_picker): detect mapped-provider auth-store credentials 2026-04-24 05:20:05 -07:00
models.py fix(copilot): wire live /models max_prompt_tokens into context-window resolver 2026-04-24 05:09:08 -07:00
nous_subscription.py
pairing.py fix(pairing): handle null user_name in pairing list display 2026-04-23 02:34:11 -07:00
platforms.py feat(cron): honor hermes tools config for the cron platform (#14798) 2026-04-23 15:48:50 -07:00
plugins_cmd.py
plugins.py docs(plugins): correct pre_gateway_dispatch doc text and add hooks.md section 2026-04-24 03:02:03 -07:00
profiles.py fix(profiles): stage profile imports to prevent directory clobbering 2026-04-23 03:02:34 -07:00
providers.py fix(providers): register alibaba-coding-plan as a first-class provider 2026-04-24 02:59:32 -07:00
runtime_provider.py fix(opencode): derive api_mode from target model, not stale config default (#15106) 2026-04-24 04:58:46 -07:00
setup.py feat: add Xiaomi MiMo v2.5-pro and v2.5 model support (#14635) 2026-04-23 10:06:25 -07:00
skills_config.py
skills_hub.py
skin_engine.py fix(skins): don't inherit status_bar_* into light-mode skins 2026-04-22 13:20:02 -07:00
status.py fix: validate nous auth status against runtime credentials 2026-04-24 05:20:05 -07:00
timeouts.py
tips.py fix(tests): resolve 17 persistent CI test failures (#15084) 2026-04-24 03:46:46 -07:00
tools_config.py chore(spotify): gate toolset off by default, add to hermes tools UI 2026-04-24 05:20:38 -07:00
uninstall.py
voice.py fix(tui): ignore SIGPIPE so stderr back-pressure can't kill the gateway 2026-04-23 16:18:15 -07:00
web_server.py feat(dashboard): reskin extension points for themes and plugins (#14776) 2026-04-23 15:31:01 -07:00
webhook.py