geekradar
Int
Explore GeekRadar

More to explore

Good gear All topics About us Say hello
GeekRadar / The comparison desk
COMPARE.
AI models
01.ai

Yi 6B

AI model Released

Context window
4,096 tokens*
Total parameters
6.06 B*
Open weights
Yes*

Capability & preference

General capability · ECI

A cross-benchmark capability estimate. Index points, not a percentage; intervals show uncertainty.

Yi 6B 94.9* Yi 6B · 95% interval 84.5–103.7

Run it locally

Inference · memory estimate

Yi 6B

6.06 B* total parameters · llama* · BF16*

Published checkpoint tensor payload: 11.3* GiB. This includes stored copies and buffers; it is separate from the runtime estimates below.

Estimated accelerator memory4.4* GiB
Weights
3.1* GiB
KV cache
0.3* GiB
Runtime allowance
1.0* GiB

One user, full model in accelerator memory.

GPUs with memory headroom

Local AI systems with memory headroom

Shared-memory systems use a planning budget of 75% of installed memory, leaving 25% for the OS and other allocations. This is an assumption, not a measured GPU allocation. Firmware and driver limits may be lower and may require configuration. Memory matches do not verify a runtime, quantized model artifact or inference speed.

Includes FP16 KV cache, prompt plus generated tokens, quantization metadata and a runtime allowance. GPU matches reserve 10% of VRAM. Memory fit is not a speed or software-compatibility guarantee; quantized weights and runtime support must be available.

Capabilities

Context window
4,096 tokens*
Open weights
Yes*
Model license
apache-2.0*
Total parameters
6.06 B*
Checkpoint tensor payload
11.29 GiB*

Input modalities

  • text*

Output modalities

  • text*

Processing & platform

Architecture
llama*

System configuration

Detailed specifications

Specifications 1 · 4 fields
Access model
Open weights (restricted use)*
Open weights
Yes*
Organization
01.AI*
Organization country
China*
Specifications 2 · 1 fields
Checkpoint memory architecture
Unique parameters
6,061,035,520*
Stored tensor elements
6,061,035,520*
Checkpoint tensor bytes
12,122,071,040*
Checkpoint data type
BF16*
Architecture
llama*
Context limit
4,096*
Transformer layers
32*
Key/value heads
4*
Attention head dimension
128*
Cache layout
full-attention*
Specifications 3 · 20 fields
Architecture
llama*
Activation function
silu*
Hidden dimension
4,096*
Feed-forward dimension
11,008*
Attention heads
32*
Transformer layers
32*
Key/value heads
4*
RMS normalization epsilon
1.0e-5*
RoPE base
5,000,000.0*
Tied word embeddings
No*
Checkpoint data type
bfloat16*
Vocabulary size
64,000*
Context window
4,096* tokens
Checkpoint input modalities
  • text*
Checkpoint licence
apache-2.0*
Checkpoint output modalities
  • text*
Total parameters
6.06103552* B
Checkpoint tensor bytes
12,122,071,040* bytes
Saved tensor elements
6,061,035,520* elements
Checkpoint tensor payload
11.289558* GiB

* Provisional data