Moonshot AI
Kimi Linear 48B A3B
Actual
65.1KDescargas562Me gustaOct 2025Publicado1.0M tokensContextoMITLicencia76 FuerteCalidad
Kimi Linear 48B A3B (48B parameters) requires approximately 33.2 GB of VRAM with Q4_K_M quantization. For the best balance of quality and speed, we recommend hardware with at least 39 GB of VRAM.
Comenzar
— copia y pega para ejecutar en localCopy-paste commands to run Kimi Linear 48B A3B on your machine.
Run
docker run --rm -it ghcr.io/ggerganov/llama.cpp:full \
--hf-repo "moonshotai/Kimi-Linear-48B-A3B-Instruct" \
--hf-file "Kimi-Linear-48B-A3B-Instruct-Q4_K_M.gguf" \
-c 4096 -ngl 99Quick specs
Parameters48B
Architecturelinear
Context1.0M tokens
Modalitytext
Min RAM18.7 GB
Rec. RAM29.3 GB (Q4_K_M)
LicenseMIT
FamilyKimi Linear
✓ Code✓ Chat✓ Reasoning
About this model
- •48B total params with 3B activated and 1M context
- •Kimi Delta Attention for lower KV-cache usage
- •Designed for efficient long-context inference and high decode throughput
Selecciones rápidas
Mejor hardware
Mejores opciones para Kimi Linear 48B A3B
Ejecutar este modelo
Opciones de cuantización
Estimaciones de VRAM por nivel de cuantización
No hardware detected — fit column shows raw VRAM estimates
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 18.7 GB | Low | — |
Q3_K_S | 3 | 23.5 GB | Low | — |
NVFP4 | 4 | 26.9 GB | Medium | — |
Q4_K_M | 4 | 29.3 GB | Medium | — |
Q5_K_M | 5 | 34.6 GB | High | — |
Q6_K | 6 | 39.4 GB | High | — |
Q8_0 | 8 | 51.4 GB | Very High | — |
F16 | 16 | 98.4 GB | Maximum | — |
Quality benchmarks
Kimi Linear 48B A3B benchmark scores
Reasoning
MMLU-Pro51.0%
GPQA Diamond—
MATH-500—
ARC Challenge—
Source: official · 2025-10-30
Compatibilidad de hardware
Estimaciones de encaje en todo el hardware
Computing compatibility...
Desglose de memoria
Reference: RTX 2060 6GB
Weights29.3 GB
KV Cache0.9 GB
Runtime2.4 GB
Headroom0.6 GB
Preguntas frecuentes
FAQ — Kimi Linear 48B A3B
Ver también