From 1cab66b89893826242ed5c08c26d9564251a844a Mon Sep 17 00:00:00 2001
From: Simon <10131203+gaomeng1900@users.noreply.github.com>
Date: Fri, 10 Jul 2026 02:08:14 -0700
Subject: [PATCH] feat(llms): update model list and patches (#611)

* fix(llms): gpt patch

* feat(llms): patch `hunyuan` models

* feat(llms): update model list
---
 packages/llms/src/models.live.test.ts         |  9 +++++--
 packages/llms/src/utils.ts                    | 27 ++++++++++++++-----
 .../src/pages/docs/features/models/page.tsx   |  6 ++++-
 3 files changed, 33 insertions(+), 9 deletions(-)

diff --git a/packages/llms/src/models.live.test.ts b/packages/llms/src/models.live.test.ts
index cb21352..467ceae 100644
--- a/packages/llms/src/models.live.test.ts
+++ b/packages/llms/src/models.live.test.ts
@@ -51,6 +51,9 @@ const MODEL_GROUPS: Record<string, string[]> = {
 		'qwen3-max',
 	],
 	OpenAI: [
+		'gpt-5.6-sol',
+		'gpt-5.6-terra',
+		'gpt-5.6-luna',
 		'gpt-5.5',
 		'gpt-5.4',
 		'gpt-5.4-mini',
@@ -80,7 +83,8 @@ const MODEL_GROUPS: Record<string, string[]> = {
 		'claude-haiku-4-5',
 	],
 	MiniMax: ['MiniMax-M3', 'MiniMax-M2.7', 'MiniMax-M2.5'],
-	xAI: ['grok-4.3', 'grok-build-0.1'],
+	xAI: ['grok-4.5', 'grok-4.3', 'grok-build-0.1'],
+	Tencent: ['hy3'],
 	MoonshotAI: ['kimi-k2.7-code', 'kimi-k2.6', 'kimi-k2.5'],
 	'Z.AI': ['glm-5.2', 'glm-5.1', 'glm-5', 'glm-4.7'],
 }
@@ -98,6 +102,7 @@ const OPENROUTER_VENDOR_SLUG: Record<string, string> = {
 	Anthropic: 'anthropic',
 	MiniMax: 'minimax',
 	xAI: 'x-ai',
+	Tencent: 'tencent',
 	MoonshotAI: 'moonshotai',
 	'Z.AI': 'z-ai',
 }
@@ -107,7 +112,7 @@ const OPENROUTER_VENDOR_SLUG: Record<string, string> = {
  * `<vendor-slug>/<lowercased-name>` heuristic — dated snapshots, "-preview"
  * suffixes, "v"-prefixed versions, or dots instead of hyphens in the
  * version number. Verified against `GET https://openrouter.ai/api/v1/models`
- * on 2026-07-03; re-check when models are added to `MODEL_GROUPS`.
+ * on 2026-07-10; re-check when models are added to `MODEL_GROUPS`.
  */
 const OPENROUTER_ID_OVERRIDES: Record<string, string> = {
 	'qwen3.6-max': 'qwen/qwen3.6-max-preview',
diff --git a/packages/llms/src/utils.ts b/packages/llms/src/utils.ts
index 0d35376..cfebb81 100644
--- a/packages/llms/src/utils.ts
+++ b/packages/llms/src/utils.ts
@@ -62,17 +62,26 @@ export function modelPatch(body: Record<string, any>, baseURL?: string) {
 	if (modelName.startsWith('gpt')) {
 		if (modelName.startsWith('gpt-5')) {
 			body.verbosity = 'low'
-
-			// gpt-5 gpt-5-mini gpt-5-nano only supports "minimal";
-			// 5.1+ supports "none".
-			body.reasoning_effort = /^gpt-5(-|$)/.test(modelName) ? 'minimal' : 'none'
-			debug(`Patch GPT-5: verbosity=low, reasoning_effort=${body.reasoning_effort}`)
 		}
 
+		// Since gpt-5.4, /chat/completions rejects any explicit reasoning_effort
+		// when function tools are present. Newer models are expected to follow.
+		// - gpt-5.1 / gpt-5.2 can fully disable reasoning
+		// - gpt-5 / -mini / -nano bottom out at "minimal"
+		// - everything else (gpt-4.x, chat-latest, gpt-5.4+) must not receive it
 		if (modelName.includes('chat-latest')) {
-			debug('Omitting reasoning_effort and temperature for chat-latest')
+			debug('Patch chat-latest: omit reasoning_effort and temperature')
 			delete body.reasoning_effort
 			delete body.temperature
+		} else if (/^gpt-5[12](-|$)/.test(modelName)) {
+			debug('Patch GPT-5.1/5.2: reasoning_effort=none')
+			body.reasoning_effort = 'none'
+		} else if (/^gpt-5(-|$)/.test(modelName)) {
+			debug('Patch GPT-5: reasoning_effort=minimal')
+			body.reasoning_effort = 'minimal'
+		} else {
+			debug('Patch GPT: omit reasoning_effort')
+			delete body.reasoning_effort
 		}
 	}
 
@@ -125,6 +134,12 @@ export function modelPatch(body: Record<string, any>, baseURL?: string) {
 		body.thinking = { type: 'disabled' }
 	}
 
+	if (modelName.startsWith('hy')) {
+		debug('Patch Hunyuan: disable thinking, reasoning_effort=low')
+		body.thinking = { type: 'disabled' }
+		body.reasoning_effort = 'low'
+	}
+
 	if (modelName.startsWith('grok')) {
 		if (/^grok-4-?3/.test(modelName)) {
 			debug('Patch Grok 4.3: reasoning_effort=none')
diff --git a/packages/website/src/pages/docs/features/models/page.tsx b/packages/website/src/pages/docs/features/models/page.tsx
index afaf4cb..b47dbca 100644
--- a/packages/website/src/pages/docs/features/models/page.tsx
+++ b/packages/website/src/pages/docs/features/models/page.tsx
@@ -29,6 +29,9 @@ const MODEL_GROUPS: Record<string, string[]> = {
 		// 'qwen3-coder-next', // low success rate
 	],
 	OpenAI: [
+		'gpt-5.6-sol',
+		'gpt-5.6-terra',
+		'gpt-5.6-luna',
 		'gpt-5.5',
 		'gpt-5.4',
 		'gpt-5.4-mini',
@@ -63,7 +66,8 @@ const MODEL_GROUPS: Record<string, string[]> = {
 		'MiniMax-M2.7',
 		'MiniMax-M2.5',
 	],
-	xAI: ['grok-4.3', 'grok-build-0.1'],
+	xAI: ['grok-4.5', 'grok-4.3', 'grok-build-0.1'],
+	Tencent: ['hy3'],
 	MoonshotAI: ['kimi-k2.7-code', 'kimi-k2.6', 'kimi-k2.5'],
 	'Z.AI': ['glm-5.2', 'glm-5.1', 'glm-5', 'glm-4.7'],
 }