Llama-3.3-70B-InstructvsKimi-K2.5
Decision Card metrics — latency, license, lineage overlap, benchmarks, and repo health from the live graph.
Model Ameta-llama
Llama-3.3-70B-Instruct
2.9Kgraph score
Open model page →Model BmoonshotaiKimi-K2.5
2.8Kgraph score
Open model page →Decision metrics
| Attribute | Llama-3.3-70B-Instruct | Kimi-K2.5 |
|---|---|---|
| License | unknown | unknown |
| ParametersModel size | 70B | — |
| Context window | — | — |
| Architecture | text-generation | image-text-to-text |
| Hardware class | — | — |
| Latency | — | — |
| $ / 1M tokens | — | — |
| Verified TTFT | — | — |
| Repo health (RIS) | — | — |
| Graph trust | 88 | 88 |
| Lineage | No lineage links | No lineage links |
| Graph score | 2.9K | 2.8K |
Decision cards
Llama-3.3-70B-Instruct
Licenseunknown
Parameters70B
Context—
Architecturetext-generation
Lifecycle
Base: meta-llama/Llama-3.1-70B
Lineage
BenchmarksPending verified runs
Repo healthNo linked repo
Economics—Economics pending verified runs
Kimi-K2.5
Licenseunknown
Parameters—
Context—
Architectureimage-text-to-text
Lifecycle
Lineage
BenchmarksPending verified runs
Repo healthNo linked repo
Economics—Economics pending verified runs
Llama-3.3-70B-Instruct
Hugging Face model with 2867 likes. Tags: transformers, safetensors, llama, text-generation, facebook, meta, pytorch, llama-3, conversational, en
Kimi-K2.5
Hugging Face model with 2827 likes. Tags: transformers, safetensors, kimi_k25, feature-extraction, compressed-tensors, image-text-to-text, conversational, custom_code, arxiv:2602.02276, license:other