chore(pricing): Update vertex-ai pricing

[siddharthsambharia-portkey]

🔄 Pricing Update: vertex-ai

📊 Summary (complete_diff mode)

Change Type	Count
➕ Models added	3
🔄 Models updated (merged)	14

➕ New Models

• gemini-2.5-pro-tts
• gemini-2.5-flash-tts
• veo-3.1-lite-generate-001

🔄 Updated Models

• gemini-3.1-pro-preview
• gemini-3-pro-image-preview
• gemini-3.1-flash-image-preview
• gemini-3.1-flash-lite-preview
• gemini-3-flash-preview
• gemini-2.5-pro
• veo-3.1-generate-001
• veo-3.0-generate-001
• veo-3.0-fast-generate-001
• text-embedding-005
• text-embedding-large-exp-03-07
• text-multilingual-embedding-002
• textembedding-gecko
• multimodalembedding

Model-to-Pricing-Page Mapping

Model ID	Publisher / Section	Source	Notes
gemini-3.1-pro-preview	Google – Gemini 3.1	API	Standard (≤200K) rates; Gemini 3 search $14/1000
gemini-3-pro-preview	Google – Gemini 3	API	Same pricing as Gemini 3.1 Pro ($2/$12/M)
gemini-3-pro-image-preview	Google – Gemini 3	API	Image output token $120/M; batch included
gemini-3.1-flash-image-preview	Google – Gemini 3.1	API	Image output token $60/M; batch included
gemini-3.1-flash-lite-preview	Google – Gemini 3.1	API	Input $0.25/M, output $1.50/M
gemini-3-flash-preview	Google – Gemini 3	API	Input $0.50/M, output $3/M
gemini-2.5-pro	Google – Gemini 2.5	API	Standard (≤200K) rates; long-context tier noted
gemini-2.5-flash	Google – Gemini 2.5	API	Standard rates
gemini-2.5-flash-preview-09-2025	Google – Gemini 2.5	API	Matches gemini-2.5-flash pricing
gemini-2.5-flash-lite	Google – Gemini 2.5	API	Input $0.10/M, output $0.40/M
gemini-2.5-flash-lite-preview-09-2025	Google – Gemini 2.5	API	Matches gemini-2.5-flash-lite pricing
gemini-2.5-flash-image	Google – Gemini 2.5	API	Image output token $30/M; batch included
gemini-2.5-computer-use-preview-10-2025	Google – Gemini 2.5	API	Matched Computer Use row; same pricing as 2.5 Pro
gemini-2.0-flash-001	Google – Gemini 2.0	API	Canonical ID with -001 suffix per ID rules
gemini-2.0-flash-lite-001	Google – Gemini 2.0	API	Canonical ID with -001 suffix per ID rules
gemini-2.5-pro-tts	Google – Gemini 2.5	API – price not found	TTS model; no pricing row found; added with price 0
gemini-2.5-flash-tts	Google – Gemini 2.5	API – price not found	TTS model; no pricing row found; added with price 0
imagen-4.0-ultra-generate-001	Google – Imagen	API	$0.06/image
imagen-4.0-generate-001	Google – Imagen	API	$0.04/image
imagen-4.0-fast-generate-001	Google – Imagen	API	$0.02/image
imagen-3.0-generate-002	Google – Imagen	API	$0.04/image
imagen-3.0-capability-001	Google – Imagen	API	Capability model mapped to imagen-3.0-generate pricing ($0.04/image)
imagen-3.0-capability-002	Google – Imagen	API	Capability model mapped to imagen-3.0-generate pricing ($0.04/image)
veo-3.1-generate-001	Google – Veo	API	$0.40/sec (Video+Audio 720p/1080p); duration=8s, count=1
veo-3.1-fast-generate-001	Google – Veo	API	$0.15/sec (Video+Audio 720p/1080p)
veo-3.1-lite-generate-001	Google – Veo	API	$0.05/sec (Video+Audio 720p)
veo-3.0-generate-001	Google – Veo	API	$0.40/sec (Video+Audio 720p/1080p)
veo-3.0-fast-generate-001	Google – Veo	API	$0.15/sec (Video+Audio)
veo-2.0-generate-001	Google – Veo	API	$0.50/sec (Video 720p)
gemini-embedding-001	Google – Embedding	API	$0.00015/1K tokens (online)
text-embedding-005	Google – Embedding	API	$0.000025/1K chars → converted to $0.0001/1K tokens
text-embedding-large-exp-03-07	Google – Embedding	API	Same pricing as text-embedding-005 (no dedicated row)
text-multilingual-embedding-002	Google – Embedding	API	$0.000025/1K chars → converted to $0.0001/1K tokens
textembedding-gecko	Google – Embedding	API	Legacy; uses standard text embedding pricing
multimodalembedding	Google – Embedding	API	Text component: $0.0002/1K chars → $0.0001/1K tokens
gemini-embedding-2-preview	Google – Embedding	API – price not found	Gemini Embedding 2 Unified Multimodal (Preview); added with price 0 pending confirmed pricing
claude-opus-4-6	Anthropic – Claude	API	@default stripped; Global pricing $5/$25/M; cache included
claude-sonnet-4-6	Anthropic – Claude	API	@default stripped; Global pricing $3/$15/M; cache included
claude-opus-4-5@20251101	Anthropic – Claude	API	Pinned version; Global pricing $5/$25/M; cache included
claude-sonnet-4-5@20250929	Anthropic – Claude	API	Pinned version; Global pricing $3/$15/M; cache included
claude-haiku-4-5@20251001	Anthropic – Claude	API	Pinned version; Global pricing $1/$5/M; cache included
claude-opus-4-1@20250805	Anthropic – Claude	API	Pinned version; Uniform pricing $15/$75/M; cache included
claude-opus-4@20250514	Anthropic – Claude	API	Pinned version; Uniform pricing $15/$75/M; cache included
claude-sonnet-4@20250514	Anthropic – Claude	API	Pinned version; Uniform pricing $3/$15/M; cache included
gpt-oss-120b-maas	OpenAI	API	MaaS model; $0.09/$0.36/M; batch included
llama-3.3-70b-instruct-maas	Meta – Llama	API	Llama 3.3 70B row; $0.72/$0.72/M; batch included
llama-4-maverick-17b-128e-instruct-maas	Meta – Llama	API	Llama 4 Maverick row; $0.35/$1.15/M; batch included
mistral-small-2503	Mistral	API	Mistral Small 3.1 row; $0.10/$0.30/M
mistral-medium-3	Mistral	API	$0.40/$2.00/M
codestral-2	Mistral	API	Codestral 2 row; $0.30/$0.90/M
deepseek-r1-0528-maas	DeepSeek	API	DeepSeek-R1 0528 row; $1.35/$5.40/M; batch included
deepseek-v3.1-maas	DeepSeek	API	DeepSeek-V3.1 row; $0.60/$1.70/M; batch included
deepseek-v3.2-maas	DeepSeek	API	DeepSeek-V3.2 row; $0.56/$1.68/M; batch included
qwen3-235b-a22b-instruct-2507-maas	Qwen	API	Qwen3-235B row; $0.22/$0.88/M; batch included
qwen3-coder-480b-a35b-instruct-maas	Qwen	API	Qwen3-Coder-480B row; $0.22/$1.80/M; batch included
qwen3-next-80b-a3b-instruct-maas	Qwen	API	Qwen3-Next-80B-Instruct row; $0.15/$1.20/M
qwen3-next-80b-a3b-thinking-maas	Qwen	API	Qwen3-Next-80B-Thinking row; $0.15/$1.20/M
kimi-k2-thinking-maas	Kimi / Moonshot	API	Kimi-K2-Thinking row; $0.60/$2.50/M
minimax-m2-maas	MiniMax	API	MiniMax-M2 row; $0.30/$1.20/M
glm-4.7-maas	ZAI.org – GLM	API	GLM-4.7 row; $0.60/$2.20/M
glm-5-maas	ZAI.org – GLM	API	GLM-5 row; $1.00/$3.20/M

Excluded Models (not in output)

Model	Publisher	Reason
gemini-live-2.5-flash-native-audio	Google	Global exclude: *-live-* streaming
jamba-large-1.6	AI21	has_deploy: true without -maas (self-deploy)
clip-vit-base-patch32, openclip	OpenAI	Non-generative (CLIP vision)
whisper-large	OpenAI	Audio transcription (global exclude)
gpt-oss	OpenAI	has_deploy: true without -maas (self-deploy)
sam3, faster-r-cnn, mask-r-cnn, retinanet	Meta	Non-generative CV/segmentation
xlm-roberta-large, roberta-large	Meta	Non-generative NLP
llama-guard, prompt-guard	Meta	Guard models (global exclude)
llama2, llama3, llama3_1, llama3-2, llama3-3, llama4, codellama-7b-hf, llama-2-quantized	Meta	has_deploy: true without -maas (self-deploy)
nllb, imagebind	Meta	Non-generative (translation / multimodal understanding)
codestral-2501-self-deploy	Mistral	has_deploy: true + "self-deploy" in name
mistral-ocr-2505	Mistral	OCR model (global exclude)
ministral-3, mistral-large-3	Mistral	has_deploy: true without -maas (self-deploy)
mistral (mistral-ai), mixtral	Mistral	has_deploy: true without -maas (self-deploy)
deepseek-r1, deepseek-v3, deepseek-v3-1, deepseek-v3-2	DeepSeek	has_deploy: true without -maas (self-deploy)
deepseek-ocr, deepseek-ocr-2, deepseek-ocr-maas	DeepSeek	OCR model (global exclude)
qwq, qwen3, qwen3-5, qwen3-vl, qwen2, qwen3-embedding, qwen3-coder, qwen3-next, qwen3-coder-next	Qwen	has_deploy: true without -maas (self-deploy)
qwen-image	Qwen	Explicit policy exclusion
kimi-k2, kimi-k2-5	Kimi	has_deploy: true without -maas (self-deploy)
minimax-m2	MiniMax	has_deploy: true without -maas (self-deploy)
glm-4.7, glm-5, glm-4.5	ZAI.org	has_deploy: true without -maas (self-deploy)
glm-image	ZAI.org	Explicit policy exclusion
glm-ocr	ZAI.org	OCR model (global exclude)
Various non-generative Google models	Google	Non-generative ML/CV/NLP (imagegeneration, virtual-try-on, bert, gemma, PaLM, etc.)

Pricing-Page-Only Models (not added — not in API)

Model	Publisher	Notes
Gemini 1.5 Flash / 1.5 Pro / 1.0 Pro	Google	On pricing page but NOT returned by get_vertex_models
Llama 3.1 405B / Llama 4 Scout	Meta	On pricing page but NOT returned by get_vertex_models
gpt-oss-20b	OpenAI	On pricing page but NOT returned by get_vertex_models
Lyria 3 Pro / Lyria 3 / Lyria 2	Google	On pricing page; excluded per global lyria-* rule (music generation)

---

Generated by Pricing Agent on 2026-04-04