Yi Models by 01.AI
About
The Yi family of large language models (LLMs), developed by 01.AI, are sophisticated language and multimodal models known for their comprehensive functionality across various dimensions 1812. Built on foundational models with 6 billion and 34 billion parameters, they incorporate advancements such as chat capabilities, models with 200K token context windows, depth-upscaled models, and vision-language processing 1812. The Yi models are designed to be bilingual in English and Chinese, achieving top performance in open-source model rankings for language understanding, commonsense reasoning, reading comprehension, and code generation 127. Available on platforms like Hugging Face and ModelScope, these models support academic research and commercial applications, aided by their open-source nature and 01.AI's focus on high-quality data engineering 12. The series also includes the multimodal Yi-VL models, excelling at tasks that involve both text and images 34, as well as the Yi-Coder models, which are optimized for code generation, editing, and long-context comprehension 9.
Current Variants
Use-when guidance is derived from seed capabilities, context, release, and replacement fields.
Use when the workload needs 200k context and 200B parameters.
Use when the workload needs 4k context, 34B parameters, and structured outputs.
Use when the workload needs 4k context and 34B parameters.
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Yi Medium | Use when the workload needs 16k context. | 2024-11 | 16k context | Current |
| Yi Large Turbo | Use when the workload needs 16k context. | 2024-01 | 16k context | Current |
| Yi Large RAG | Use when the workload needs 16k context. | 2024-01 | 16k context | Current |
| Yi Large FC | Use when the workload needs 32k context. | 2024-01 | 32k context | Current |
| Yi Lightning | Use when the workload needs 16k context. | 2024-01 | 16k context | Current |
| Yi Lightning Lite | Use when the workload needs 16k context. | 2024-01 | 16k context | Current |
| Yi Spark | Use when the workload needs 16k context. | 2024-01 | 16k context | Current |
| Yi Medium 200K | Use when the workload needs 200k context and 200B parameters. | 2024-01 | 200k context200B parameters | Current |
| Together AI Yi-34B-Chat | Use when the workload needs 4k context, 34B parameters, and structured outputs. | 2024-01 | 4k context34B parametersstructured outputs | Current |
| Fireworks Yi-34B-Chat | Use when the workload needs 4k context and 34B parameters. | 2024-01 | 4k context34B parameters | Current |
Release Timeline
2 release groupsSpecifications(11 models)
| Model | Released | Context | Parameters | Structured Outputs |
|---|---|---|---|---|
| Yi Medium | 2024-11 | 16k | — | No |
| Yi Large Turbo | 2024-01 | 16k | — | No |
| Yi Large RAG | 2024-01 | 16k | — | No |
| Yi Large FC | 2024-01 | 32k | — | No |
| Yi Lightning | 2024-01 | 16k | — | No |
| Yi Lightning Lite | 2024-01 | 16k | — | No |
| Yi Spark | 2024-01 | 16k | — | No |
| Yi Medium 200K | 2024-01 | 200k | 200B | No |
| Together AI Yi-34B-Chat | 2024-01 | 4k | 34B | Yes |
| Fireworks Yi-34B-Chat | 2024-01 | 4k | 34B | No |
Available From(3 providers)
Pricing
| Model | Provider | Input / 1M | Output / 1M | Type |
|---|---|---|---|---|
| Together AI Yi-34B-Chat | Together AI | $0.3 | $0.3 | Serverless |
| Fireworks Yi-34B-Chat | Fireworks AI | $0.4 | $0.4 | Serverless |
Frequently Asked Questions
- What is Yi used for?
- Yi is used for structured outputs, coding, and chatbot and role-playing use cases. The family description and listed model capabilities point to those workloads as the best fit.
- How does Yi compare to Claude 4.8?
- Yi by 01.AI is strongest where you need structured outputs, while Claude 4.8 by Anthropic is the closest related family to check for vision and multimodal work. Yi has 11 listed variants and reaches up to 200k context, while Claude 4.8 reaches up to 1m context, so compare the specs and pricing tables before choosing a production model.
- Which Yi model should I use?
- For the lowest listed input price, start with Together AI Yi-34B-Chat through Together AI at $0.3/1M input tokens. For the most capable/latest local choice, evaluate Together AI Yi-34B-Chat with 4k context and structured outputs.
