Yi-34B
Parameters
34B
Context Length
4K
Modality
Text
Architecture
Dense
KV Cache Rate
240 KB/tok
License
Apache 2.0
Release Date
2 Nov 2023
Knowledge Cutoff
Jun 2023
API Pricing (per 1M)
Input: $0.90 · Output: $0.90
Hardware Sizing
Precision
Context
Workload Memory: 19.27 GB
INT4 · 4K context
Base Weights: 15.83 GB
KV Cache: 0.94 GB
Runtime & Graphs: 2.50 GB
Minimum Required Hardware
1x NVIDIA L4 24GB
19.3 / 21.6 GB usable
2x RTX 4090 24GB
19.3 / 43.2 GB usable · TP 2
Apple M3 Pro 36GB
19.3 / 27.0 GB usable
Determine requirements for your specific workload:
Open in VRAM Calculator
Architecture Diagram
Evaluation Benchmarks
Rank
#233
| Benchmark | Score | Rank |
|---|---|---|
General Text | 1184 | 175 |
Rankings
Overall Rank
#233
Coding Rank
-
About Yi-34B
Yi-34B is a foundational bilingual model developed by 01.AI, pretrained on 3T multilingual tokens for advanced English and Chinese text comprehension. It features Grouped-Query Attention and supports context lengths extending up to 200K tokens.
Technical Specifications
Attention
Attention Structure
Multi-Head Attention
Attention Heads
56
Key-Value Heads
8
Position Embedding
Absolute Position Embedding
RoPE Theta
5,000,000
Normalization
RMS Normalization
Activation Function
SwigLU
Dimensions
Hidden Dimension Size
7,168
Number of Layers
60
FFN Intermediate Size (Dense)
20,480
Tokenizer
Vocabulary Size
64,000