LFM2.5-ColBERT-350M
A late-interaction multilingual retrieval model that uses per-token vectors matched via MaxSim, built on LFM2.5-350M-Base.
Fact sheet
Parameters
Context length
Architecture
Shape
Weights stored as
Task
Languages
Built on
Released
Last updated
Downloads (last 30 days)
Licence and access
Licence
Access
Security
Hugging Face security scan
Weight files by format
Facthory’s checks
Every weight file is safetensors or GGUF.
No file flagged.
The card declares its licence.
The card reports no safety evaluation.
GPU calculator
How much memory this model needs to serve at a quantization, context and number of concurrent sequences, which GPUs hold it, and how fast it can generate. Estimates from the model’s published size and shape, with the assumptions below.
Weights
Context per sequence: 8K tokens
Concurrent sequences: 1
KV cache
Memory
1.1 GB
Weights: 0.7 GB
KV cache: 0.3 GB
Runtime: 0.1 GB
| GPU | Memory | GPUs needed | Per sequence | All sequences |
|---|---|---|---|---|
| NVIDIA L4 | 24 GB | 1 × | 185 tokens/s | 185 tokens/s |
| NVIDIA L40S | 48 GB | 1 × | 532 tokens/s | 532 tokens/s |
| NVIDIA RTX 6000 Ada | 48 GB | 1 × | 591 tokens/s | 591 tokens/s |
| NVIDIA RTX PRO 6000 Blackwell | 96 GB | 1 × | 1,103 tokens/s | 1,103 tokens/s |
| NVIDIA A100 80GB SXM | 80 GB | 1 × | 1,255 tokens/s | 1,255 tokens/s |
| NVIDIA H100 SXM | 80 GB | 1 × | 2,061 tokens/s | 2,061 tokens/s |
| NVIDIA H200 SXM | 141 GB | 1 × | 2,954 tokens/s | 2,954 tokens/s |
| AMD Instinct MI300X | 192 GB | 1 × | 3,261 tokens/s | 3,261 tokens/s |
| AMD Instinct MI325X | 256 GB | 1 × | 3,692 tokens/s | 3,692 tokens/s |
Assumptions
- Runtime memory (activations, buffers) is 10% of weights and KV cache, plus 1 GB per GPU for the framework.
- A GPU is filled to at most 90% of its memory; tensor parallelism uses 1, 2, 4 or 8 GPUs in one server.
- Generation is limited by memory bandwidth at 60% of the vendor’s peak: each step reads the weights once and every sequence’s KV cache. A mixture-of-experts model reads only its active experts, so it runs faster than shown.
- GPU memory and bandwidth are the vendors’ published figures.
Profile
Kind
Lineage
Family
Uses
In and out
Written on 12 Nov 2026 by Workers AI (@cf/zai-org/glm-4.7-flash) from the model card and the hub’s metadata: a reading of the publisher’s text, not checked by Facthory.
Published benchmarks
The model card publishes no benchmark results.
EU AI Act
What the EU AI Act asks of a general-purpose AI model depends on who places it on the EU market and how. These facts bear on it:
- Licence: other
- Access: Weights download without approval
- Parameters: 353.3M
General-purpose AI model (Article 3(63))
Not indicated. Its task (sentence-similarity) generates no text, audio, images or video, which the Commission’s indicative criterion asks for.
Articles 51 to 55, the obligations for general-purpose AI models, do not bind its provider.
Facthory’s EU AI Act score appears once a person has reviewed these checks. The articles we score against
