← Back to the Model Guide
Cheapest model / job
RAG and retrieval: the cheapest models
Ground answers in your own documents.
How this page ranks
- Ranked by blended price per million tokens: three parts input to one part output, at the OpenRouter list price for the model. Lowest first. Ties go to the lower input price.
- A model counts for this job when its taxonomy v2 classification names rag and retrieval, or its old category maps to it.
- Each model must list text input on OpenRouter or on its own record.
- Prices and scores come from the market tables only, each with the day it is for. The price on a model's own record is not used.
- Missing data never ranks. A model with no confirmed price in the last 30 days, a price in another unit, a $0 listing or unknown inputs and outputs is listed apart, with the reason.
Market data as of 10 Nov 2026. Prices: OpenRouter models API (official).
Ranked: 121 of 360 models
Best for rag and retrieval →| # | Model | Input / 1M | Output / 1M | Blended / 1M | Price date |
|---|---|---|---|---|---|
| 1 |  Mistral NemoMistral | $0.029 | $0.030 | $0.029 | 10 Nov 2026 |
| 2 | SExtractScaledown | $0.040 | $0 | $0.030 | 10 Nov 2026 |
| 3 |  Granite 4.0 H MicroCf | $0.017 | $0.112 | $0.041 | 25 Oct 2026 |
| 4 | IGranite 4.0 MicroIbm Granite | $0.017 | $0.112 | $0.041 | 25 Oct 2026 |
| 5 |  Gemma 3 4BGoogle | $0.050 | $0.100 | $0.063 | 10 Nov 2026 |
| 6 |  Command R7B (12-2024)Cohere | $0.037 | $0.150 | $0.066 | 10 Nov 2026 |
| 7 | ILing 3.1 FlashInclusionai | $0.045 | $0.135 | $0.068 | 10 Nov 2026 |
| 8 | Gemma 3 12BGoogle | $0.050 | $0.150 | $0.075 | 10 Nov 2026 |
| 9 | Gemma 4 26B A4B ITCf | $0.080 | $0.100 | $0.085 | 10 Nov 2026 |
| 10 | USolar Mini 4Upstage | $0.050 | $0.200 | $0.088 | 10 Nov 2026 |
| 11 | ISchematron V2 SmallInference Net | $0.050 | $0.230 | $0.095 | 10 Nov 2026 |
| 12 |  Nemotron 3 Nano 30B A3BNVIDIA | $0.060 | $0.240 | $0.105 | 10 Nov 2026 |
| 13 |  Nova Lite 1.0Amazon | $0.060 | $0.240 | $0.105 | 10 Nov 2026 |
| 14 |  DeepSeek V4 FlashAccounts | $0.090 | $0.180 | $0.113 | 10 Nov 2026 |
| 15 | DeepSeek V4 Flash 0423DeepSeek | $0.090 | $0.180 | $0.113 | 10 Nov 2026 |
| 16 | BSeed 1.6 FlashBytedance Seed | $0.075 | $0.300 | $0.131 | 9 Nov 2026 |
| 17 |  Llama 4 ScoutMeta | $0.100 | $0.300 | $0.150 | 10 Nov 2026 |
| 18 | Nemotron 3 SuperNVIDIA | $0.075 | $0.400 | $0.156 | 10 Nov 2026 |
| 19 | Gemini 2.5 Flash LiteGoogle | $0.100 | $0.400 | $0.175 | 10 Nov 2026 |
| 20 |  GPT-4.1 NanoOpenAI | $0.100 | $0.400 | $0.175 | 10 Nov 2026 |
| 21 | BSeed-2.0-MiniBytedance Seed | $0.100 | $0.400 | $0.175 | 10 Nov 2026 |
| 22 |  Claude Haiku 5.5Anthropic | $0.100 | $0.500 | $0.200 | 10 Nov 2026 |
| 23 | GPT-6 LunaOpenAI | $0.100 | $0.500 | $0.200 | 10 Nov 2026 |
| 24 | LLFM2.5-24B-A2BLiquid | $0.050 | $0.700 | $0.212 | 10 Nov 2026 |
| 25 |  Qwen3 30B A3BQwen | $0.120 | $0.500 | $0.215 | 10 Nov 2026 |
Showing the top 25 of 121 ranked models.
Not ranked, and why (239)
No confirmed price in the market tables in the last 30 days 238
- Agnes 1.5 Pro Not matched to an OpenRouter model
- AI21 Jamba 1.5 Mini Not matched to an OpenRouter model
- AionLabs: Aion-1.0-Mini Not matched to an OpenRouter model
- Apodex 1.1 Mini (free) Not matched to an OpenRouter model
- Arcee AI: Spotlight Not matched to an OpenRouter model
- Baidu: CoBuddy (free) Not matched to an OpenRouter model
- Baidu: Qianfan-OCR-Fast Not matched to an OpenRouter model
- bevformer Not matched to an OpenRouter model
- BGE M3 Not matched to an OpenRouter model
- BGE Reranker Base Not matched to an OpenRouter model
- BGE Small EN v1.5 Not matched to an OpenRouter model
- Bielik 11B v3.0 Instruct Not matched to an OpenRouter model
- Claude 4.6 Opus Thinking Max Not matched to an OpenRouter model
- Claude Haiku 5.5 (batch) Not matched to an OpenRouter model
- Claude Sonnet 3.5 v2 Not matched to an OpenRouter model
- Claude Sonnet 4 Not matched to an OpenRouter model
- Claude Sonnet 4.6 Thinking Not matched to an OpenRouter model
- Claude-Haiku-3.5 Not matched to an OpenRouter model
- Codestral Embed Not matched to an OpenRouter model
- Cohere Command A Not matched to an OpenRouter model
- Cohere Command R Not matched to an OpenRouter model
- Cohere Command R 08-2024 Not matched to an OpenRouter model
- Cohere Command R+ Not matched to an OpenRouter model
- Cohere Command R+ 08-2024 Not matched to an OpenRouter model
- Cohere Rerank 4 Fast Not matched to an OpenRouter model
- Cohere Rerank 4 Pro Not matched to an OpenRouter model
- Cohere: Command R Not matched to an OpenRouter model
- DeepSeek OCR Not matched to an OpenRouter model
- DeepSeek Reasoner Not matched to an OpenRouter model
- DeepSeek V4 Flash Vision Exp (batch) Not matched to an OpenRouter model
- DeepSeek V4 Pro (Thinking) Not matched to an OpenRouter model
- DeepSeek V4 Pro Cheaper Not matched to an OpenRouter model
- DeepSeek-OCR Not matched to an OpenRouter model
- deepseek/deepseek-ocr-2 Not matched to an OpenRouter model
- Deepseek/DeepSeek-V3.2 Not matched to an OpenRouter model
- Doubao-Seed-2.0-lite Not matched to an OpenRouter model
- Doubao-Seed-2.0-mini Not matched to an OpenRouter model
- Doubao-Seed-2.0-pro Not matched to an OpenRouter model
- E5 Mistral 7B Not matched to an OpenRouter model
- E5 Multi-Lingual Large Embeddings 0.6B Not matched to an OpenRouter model
- Embed v4.0 Not matched to an OpenRouter model
- ERNIE 4.5 VL 28B A3B Not matched to an OpenRouter model
- ERNIE 4.5 VL 424B A47B No confirmed price for baidu/ernie-4.5-vl-424b-a47b
- ERNIE 5.0 Not matched to an OpenRouter model
- ERNIE-4.5-21B-A3B-Thinking Not matched to an OpenRouter model
- ERNIE-4.5-VL-28B-A3B-Thinking Not matched to an OpenRouter model
- Gemini 1.5 Flash Not matched to an OpenRouter model
- Gemini 2.5 Flash (batch) Not matched to an OpenRouter model
- Gemini 3 Flash Thinking Not matched to an OpenRouter model
- Gemini 3.1 Flash Lite (batch) Not matched to an OpenRouter model
- Gemini 3.1 Pro (Preview High) Not matched to an OpenRouter model
- Gemini 3.1 Pro (Preview Low) Not matched to an OpenRouter model
- Gemini 3.5 Flash Lite (batch) Not matched to an OpenRouter model
- Gemini 3.5 Flash Thinking Not matched to an OpenRouter model
- Gemini Embedding 001 Not matched to an OpenRouter model
- Gemini Embedding 2 No confirmed price for google/gemini-embedding-2
- Gemini Flash-Lite Latest Not matched to an OpenRouter model
- Gemini Omni Flash Preview Not matched to an OpenRouter model
- Gemini-2.0-Flash Not matched to an OpenRouter model
- Gemini-2.0-Flash-Lite Not matched to an OpenRouter model
- Gemma 3n E2b It Not matched to an OpenRouter model
- Gemma 4 31B (batch) Not matched to an OpenRouter model
- Gemma 4 31B (free) Not matched to an OpenRouter model
- Gemma 4 E4B IT Not matched to an OpenRouter model
- GLM 4.5V Thinking Not matched to an OpenRouter model
- GLM 4.6 Thinking Not matched to an OpenRouter model
- GLM 4.6V Flash Not matched to an OpenRouter model
- GLM 4.6V Flash (Free) Not matched to an OpenRouter model
- GLM 4.6V Original Not matched to an OpenRouter model
- GLM 4.7 Flash (Free) Not matched to an OpenRouter model
- GLM 4.7 Original Thinking Not matched to an OpenRouter model
- GLM 5 Original Not matched to an OpenRouter model
- GLM 5 Original Thinking Not matched to an OpenRouter model
- GLM 5.1 Thinking TEE Not matched to an OpenRouter model
- GLM 5.2 (free) Not matched to an OpenRouter model
- GLM 5V Turbo Thinking Not matched to an OpenRouter model
- GLM-4.6V-Flash Not matched to an OpenRouter model
- Google: Gemini 2.0 Flash Lite Not matched to an OpenRouter model
- Google: Gemini 2.5 Flash Lite Preview 09-2025 Not matched to an OpenRouter model
- GPT-4 Turbo (batch) Not matched to an OpenRouter model
- GPT-4.1 (batch) Not matched to an OpenRouter model
- GPT-4.1 Mini (batch) Not matched to an OpenRouter model
- GPT-5 Pro (batch) Not matched to an OpenRouter model
- GPT-5.2 Pro (batch) Not matched to an OpenRouter model
- GPT-5.4 (batch) Not matched to an OpenRouter model
- GPT-5.4 Pro (batch) Not matched to an OpenRouter model
- GPT-5.5 (batch) Not matched to an OpenRouter model
- GPT-5.5 Instant Not matched to an OpenRouter model
- GPT-5.5 Pro (batch) Not matched to an OpenRouter model
- GPT-5.6 Terra (batch) Not matched to an OpenRouter model
- GPT-6 Astra (batch) Not matched to an OpenRouter model
- GPT-6 Sol (batch) Not matched to an OpenRouter model
- GPT-6.1 Sol (batch) Not matched to an OpenRouter model
- GPT-OSS-120B-CS Not matched to an OpenRouter model
- Granite 4.1 8B No confirmed price for ibm-granite/granite-4.1-8b
- Grok 3 Not matched to an OpenRouter model
- Grok 4.1 Fast Reasoning Not matched to an OpenRouter model
- Grok 4.2 Fast Not matched to an OpenRouter model
- Grok 4.2 Fast Non Reasoning Not matched to an OpenRouter model
- Grok 4.20 (Reasoning) Not matched to an OpenRouter model
- Grok 4.3 (batch) Not matched to an OpenRouter model
- Grok Code Fast 1 Not matched to an OpenRouter model
- Hermes 4 Large (Thinking) Not matched to an OpenRouter model
- Hy3 (free) Not matched to an OpenRouter model
- inclusionAI/Ling-flash-2.0 Not matched to an OpenRouter model
- Inkling (free) Not matched to an OpenRouter model
- Inkling Small (batch) Not matched to an OpenRouter model
- Interfaze Beta Not matched to an OpenRouter model
- Jamba Large 1.7 No confirmed price for ai21/jamba-large-1.7
- Jev Router Not matched to an OpenRouter model
- Kilo Auto Balanced Not matched to an OpenRouter model
- Kilo Auto Free Not matched to an OpenRouter model
- Kilo Auto Frontier Not matched to an OpenRouter model
- Kimi-K2-Instruct Not matched to an OpenRouter model
- LFM2.5-1.2B-Thinking (free) Not matched to an OpenRouter model
- LFM2.5-2.6B (free) Not matched to an OpenRouter model
- Ling 3.0 Flash Sante No confirmed price for inclusionai/ling-3.0-flash-sante
- Ling-1T Not matched to an OpenRouter model
- Llama 3.2 11B Vision Instruct No confirmed price for meta-llama/llama-3.2-11b-vision-instruct
- Llama 3.2 11B Vision Instruct Not matched to an OpenRouter model
And 118 more.
Priced per image, song or other unit, not per token 1
- Cohere Rerank 3.5 Priced per unknown
Automated from dated market data, not Enterprise DNA tested picks. Check the price and terms with the provider before you commit. Methodology: how the model numbers work.