Updated 24 min ago · first seen 11 Sept 2026
model_01M294ZGJMJ4129YY5TNSN2916
- Parameters
- 16.4B · 3B active
- T2 · 1 h ago
- Released
- 21 Jun 2025
- T2 · 24 min ago
- License
- mit
- T2 · 24 min ago
Specification
- Release date
- 21 Jun 2025
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 24 min agomedium
- Openness
- open-weights
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
- License
- mit
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 24 min agomedium
- Architecture
- KimiVLForConditionalGeneration
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 24 min agomedium
- Parameters
- 16.4B
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
- Active parameters
- 3B
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 24 min agomedium
- Base model
- moonshotai/Kimi-VL-A3B-Instruct
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 24 min agomedium
- File size
- 32.8 GB
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 24 min agomedium
- Hugging Face repo
- moonshotai/Kimi-VL-A3B-Thinking-2506
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
- Pipeline tag
- image-text-to-text
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
- Model card
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
- Downloads
- 30,016
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 24 min agomedium
- Likes
- 383
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 24 min agomedium
- Gated
- No
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 24 min agomedium
- Last modified
- 2026-01-30T04:53:11+00:00
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
- Library name
- transformers
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 24 min agomedium
- Downloads all time
- 986,237
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 24 min agomedium
- Model type
- kimi_vl
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 24 min agomedium
- Tags
- transformers, safetensors, kimi_vl, feature-extraction, image-text-to-text, custom_code
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 24 min agomedium
- Weights dtype
- BF16, F32
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 24 min agomedium
Each value shows its source, tier and observation time. Conflicting claims are kept side by side and flagged — never averaged. How AI Atlas records facts →
Provenance
Attributed facts
20
Source tiers
T220
Freshest observation
24 min ago
Conflicts
None
Modalities
Modalities unavailable.
Capabilities
Tool calling
Unavailable
Structured output
Unavailable
Reasoning
Unavailable
Vision
Unavailable
Audio
Unavailable
Fine-tuning available
Unavailable
No capability flags have been observed from a source yet — we do not infer them.
No structured attributes yet.
No benchmark results recorded
Current prices
No current prices recorded
Price history
Memory need = bytes per parameter (4-bit ≈ 0.5 × 1.15 overhead, 8-bit 1.0, fp16 2.0) + a KV-cache allowance. Not a measurement.
| Hardware | Quantization | Memory | Est. need | Fits |
|---|---|---|---|---|
| Apple M3 Ultra | 4bit | — | 9.9 GB | Yes |
| AMD Instinct MI325X | 4bit | 256 GB | 9.9 GB | Yes |
| AMD Instinct MI300X | 4bit | 192 GB | 9.9 GB | Yes |
| Apple M2 Ultra | 4bit | — | 9.9 GB | Yes |
| NVIDIA B200 | 4bit | 180 GB | 9.9 GB | Yes |
| NVIDIA H200 | 4bit | 141 GB | 9.9 GB | Yes |
| Apple M1 Ultra | 4bit | — | 9.9 GB | Yes |
| Apple M3 Max | 4bit | — | 9.9 GB | Yes |
| Apple M4 Max | 4bit | — | 9.9 GB | Yes |
| NVIDIA DGX Spark | 4bit | 128 GB | 9.9 GB | Yes |
| Apple M2 Max | 4bit | — | 9.9 GB | Yes |
| NVIDIA A100 80GB | 4bit | 80 GB | 9.9 GB | Yes |
| NVIDIA H100 SXM | 4bit | 80 GB | 9.9 GB | Yes |
| Apple M1 Max | 4bit | — | 9.9 GB | Yes |
| Apple M4 Pro | 4bit | — | 9.9 GB | Yes |
| Apple M3 Pro | 4bit | — | 9.9 GB | Yes |
| Apple M1 Pro | 4bit | — | 9.9 GB | Yes |
| Apple M2 Pro | 4bit | — | 9.9 GB | Yes |
| Apple M4 | 4bit | — | 9.9 GB | Yes |
| NVIDIA GeForce RTX 5090 | 4bit | 32 GB | 9.9 GB | Yes |
| Apple M2 | 4bit | — | 9.9 GB | Yes |
| Apple M3 | 4bit | — | 9.9 GB | Yes |
| NVIDIA GeForce RTX 3090 | 4bit | 24 GB | 9.9 GB | Yes |
| NVIDIA GeForce RTX 4090 | 4bit | 24 GB | 9.9 GB | Yes |
| Apple M1 | 4bit | — | 9.9 GB | Yes |
Ancestors 2
This model
Kimi-VL-A3B-Thinking-2506
16.4B params
Quantizations 0
None recorded.
Descendants 0
None recorded.
Papers 1
- arXiv:2504.07491Active35
Repositories 0
No repositories linked yet.
As of
Rewind the record: see this entity's attributes exactly as AI Atlas knew them on a given day.
Claim history · Openness
Opennessopenness1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| open-weights | → current | current | Hugging Face Hub (public pages, model cards, papers)T2 | medium | deterministic |
Claims are temporal and append-only: a new observation closes the previous claim (valid_to) instead of overwriting it. Conflicting claims from different sources are kept side by side and flagged — never averaged. Methodology →
New model: Kimi-VL-A3B-Thinking-2506 (Moonshot AI)
huggingface
| Source | Document | Type | Tier | Last observed | Snapshots |
|---|---|---|---|---|---|
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/moonshotai/Kimi-VL-A3B-Thinking-2506 | model_page | T2· Quality secondary | 23 min ago | 1 |
| Hugging Face Hub (public pages, model cards, papers) | huggingface.co/models?author=moonshotai&p=0&sort=downloads | listing | T2· Quality secondary | 1 h ago | 2 |
Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.