Compare commits

...

2 commits

Author SHA1 Message Date
01594f3d95
feat: add glm-4.7-flash and qwen3.5 models with provider routes and pricing 2026-02-18 00:36:08 -06:00
ba8cfb4fa2
feat(settings): add timeout_limit and configure engine weights and timeouts
BREAKING CHANGE: timeout_limit added to search section and engine configurations updated with weights and timeouts
2026-02-18 00:36:01 -06:00
2 changed files with 125 additions and 5 deletions

View file

@ -252,6 +252,15 @@ models:
model: "mistralai/devstral-medium" model: "mistralai/devstral-medium"
# ═══ TIER 6: GLM ═══ # ═══ TIER 6: GLM ═══
- name: "glm-4.7-flash"
routes:
- provider: deepinfra
model: "zai-org/GLM-4.7-Flash"
pricing: { input: 0.06, output: 0.40 }
- provider: openrouter
model: "z-ai/glm-4.7-flash"
pricing: { input: 0.06, output: 0.40 }
- name: "glm-4.6" - name: "glm-4.6"
routes: routes:
- provider: ollama-cloud - provider: ollama-cloud
@ -304,9 +313,28 @@ models:
- provider: openrouter - provider: openrouter
model: "moonshotai/kimi-k2.5" model: "moonshotai/kimi-k2.5"
# ═══ TIER 8: SiliconFlow (Qwen) ═══ # ═══ TIER 8: Qwen ═══
- name: "qwen3.5"
routes:
- provider: ollama-cloud
model: "qwen3.5:cloud"
# - provider: openrouter
# model: "qwen/qwen3.5-397b-a17b"
# pricing: { input: 0.60, output: 3.60 }
- name: "qwen3-coder" - name: "qwen3-coder"
routes: routes:
- provider: ollama-cloud
model: "qwen3-coder:480b-cloud"
- provider: deepinfra
model: "Qwen/Qwen3-Coder-480B-A35B-Instruct"
pricing: { input: 0.22, output: 1.00 }
- provider: openrouter
model: "qwen/qwen3-coder"
pricing: { input: 0.22, output: 1.00 }
- provider: cerebras
model: "qwen3-coder-480b"
pricing: { input: 2.00, output: 2.00 }
- provider: siliconflow - provider: siliconflow
model: "Qwen/Qwen3-Coder-480B-A35B-Instruct" model: "Qwen/Qwen3-Coder-480B-A35B-Instruct"
pricing: { input: 1.14, output: 2.28 } pricing: { input: 1.14, output: 2.28 }

View file

@ -10,6 +10,7 @@ search:
safe_search: 0 safe_search: 0
autocomplete: "google" autocomplete: "google"
default_lang: "en" default_lang: "en"
timeout_limit: 5.0 # wait up to 5s total; fast engines respond sooner
formats: formats:
- html - html
- json - json
@ -22,34 +23,125 @@ server:
ui: ui:
static_use_hash: true static_use_hash: true
results_on_new_tab: false
theme_args:
simple_style: auto # respects system dark/light mode
engines: engines:
# --- General Web (higher weight = results ranked higher in merged list) ---
- name: google - name: google
engine: google engine: google
shortcut: g shortcut: g
language: en language: en
region: en-US region: en-US
weight: 2.0 # best index quality
- name: duckduckgo timeout: 4.0
engine: duckduckgo
shortcut: ddg
- name: brave - name: brave
engine: brave engine: brave
shortcut: br shortcut: br
weight: 1.5 # independent index, good quality
timeout: 3.0
- name: duckduckgo
engine: duckduckgo
shortcut: ddg
weight: 1.5
timeout: 3.0
- name: bing
engine: bing
shortcut: b
weight: 1.5 # large index, complements Google well
timeout: 3.0
- name: startpage
engine: startpage
shortcut: sp
weight: 1.2 # Google results with privacy; adds diversity
timeout: 4.0
- name: qwant
engine: qwant
shortcut: q
weight: 1.0
timeout: 3.0
- name: mojeek
engine: mojeek
shortcut: mjk
weight: 0.8 # smaller independent index; good for diversity
timeout: 3.0
# --- Knowledge / Reference ---
- name: wikipedia - name: wikipedia
engine: wikipedia engine: wikipedia
shortcut: wp shortcut: wp
weight: 1.5
timeout: 2.0
# --- Tech / Code ---
- name: github - name: github
engine: github engine: github
shortcut: gh shortcut: gh
categories: [it]
weight: 1.5
timeout: 3.0
- name: stackoverflow - name: stackoverflow
engine: stackoverflow engine: stackoverflow
shortcut: so shortcut: so
categories: [it]
weight: 1.5
timeout: 3.0
- name: npm
engine: npm
shortcut: npm
categories: [it]
weight: 1.2
timeout: 2.5
- name: pypi
engine: pypi
shortcut: pip
categories: [it]
weight: 1.2
timeout: 2.5
# --- Academic ---
- name: arxiv - name: arxiv
engine: arxiv engine: arxiv
shortcut: arx shortcut: arx
categories: [science]
weight: 1.5
timeout: 3.0
- name: semantic scholar
engine: semantic scholar
shortcut: ss
categories: [science]
weight: 1.2
timeout: 3.5
# --- Community / News ---
- name: reddit
engine: reddit
shortcut: r
categories: [social media]
weight: 1.2
timeout: 3.0
- name: hackernews
engine: hackernews
shortcut: hn
categories: [it, news]
weight: 1.2
timeout: 3.0
- name: google news
engine: google news
shortcut: gn
categories: [news]
weight: 1.5
timeout: 3.0