InternScience

Agents-A1 4B

Name: Agents-A1 4B
Rating: 75 (196 reviews)
Author: InternScience

Frontera

HuggingFace

65.2KDescargas64Me gustaJul 2026Publicado262K tokensContextoApache 2.0Licencia62 BuenoCalidad

Agents-A1 4B (4.5B parameters) requires approximately 4.7 GB of VRAM with Q4_K_M quantization. For the best balance of quality and speed, we recommend hardware with at least 6 GB of VRAM.

Comenzar

— copia y pega para ejecutar en local

Copy-paste commands to run Agents-A1 4B on your machine.

Run

docker run --rm -it ghcr.io/ggerganov/llama.cpp:full \
  --hf-repo "InternScience/Agents-A1-4B" \
  --hf-file "Agents-A1-4B-Q4_K_M.gguf" \
  -c 4096 -ngl 99

Quick specs

Parameters4.5B

Architecturedense

Context262K tokens

Modalitytext+vision

Min RAM1.8 GB

Rec. RAM2.7 GB (Q4_K_M)

LicenseApache 2.0

FamilyAgents-A1

✓ Reasoning

About this model

Agents-A1-4B is the compact dense member of InternScience's Agents-A1 family — a 4.5B vision-capable agentic model distilled from the same long-horizon agent training recipe as the 35B MoE. It targets edge and laptop deployment while keeping the family's 262K context and hybrid linear/full attention backbone.

•4.5B dense and vision-capable — fits in ~3GB at Q4.
•262K context, unusually long for a 4B-class model.
•Hybrid linear + full attention (1-in-4 layers) keeps long-context KV cache small.
•Shares the Agents-A1 long-horizon agentic training recipe.

Modelos relacionados

Selecciones rápidas

Mejor económicoA

Intel Arc A380 6GB~$139 — 27 tok/s

Mejor en generalA

RTX 2060 6GB~$349 — 57 tok/s

Mejor hardware

Mejores opciones para Agents-A1 4B

Ejecutar este modelo

Agents-A1 4B on RTX 2060 6GB Agents-A1 4B on GTX 1660 Super 6GB Agents-A1 4B on RTX 3080 10GB

Opciones de cuantización

Estimaciones de VRAM por nivel de cuantización

No hardware detected — fit column shows raw VRAM estimates

Quant	Bits	VRAM	Quality	Fit
Q1_0_G128	1.125	0.6 GB	Very Low	—
Q2_0_G128	1.71	1.2 GB	Low	—
Q2_K	2	1.8 GB	Low	—
Q3_K_S	3	2.2 GB	Low	—
NVFP4	4	2.5 GB	Medium	—
Q4_K_M	4	2.7 GB	Medium	—
Q5_K_M	5	3.2 GB	High	—
Q6_K	6	3.7 GB	High	—
Q8_0	8	4.8 GB	Very High	—
F16	16	9.2 GB	Maximum	—

Quality benchmarks

Agents-A1 4B benchmark scores

Benchmark verified

Coding

SWE-bench Verified—

HumanEval+—

Aider Polyglot—

LiveCodeBench59.6%

General

Chatbot Arena—

IFEval94.8%

Source: official · 2026-07-13

Compatibilidad de hardware

Estimaciones de encaje en todo el hardware

Abrir calculadora

Computing compatibility...

Desglose de memoria

Reference: RTX 2060 6GB

Weights2.7 GB

KV Cache0.5 GB

Runtime0.9 GB

Headroom0.6 GB

Preguntas frecuentes

FAQ — Agents-A1 4B

Ver también

Guía de cuantización Metodología de puntuación Abrir calculadora