AI 모델 인텔리전스

기능 · 2026-08-14

오픈 웨이트 AI 모델

라이선스 하에 학습된 가중치를 공개하는 AI 모델 비교 — 셀프 호스팅, 파인튜닝, 규제 환경용.

이게 뭔가요?

  • 오픈 웨이트 모델은 가중치를 공개합니다(주로 Apache 2.0, MIT 또는 오픈 연구 라이선스).
  • 참고: 오픈 웨이트 ≠ 오픈 소스 — 학습 데이터와 코드는 보통 공개되지 않습니다.

왜 중요한가

  • 자체 GPU에서 셀프 호스팅, 비공개 데이터로 파인튜닝, 오프라인 운영, 규제 환경 배포가 가능합니다.
  • 표에는 가장 저렴한 API 호스트 가격이 표시됩니다 — 자체 하드웨어가 있다면 API 비용 없이 동일 모델을 운영할 수 있습니다.

이 기능을 지원하는 모델 626개

모델벤더입력 / 1M출력 / 1M컨텍스트제공자
Whisper Large v3scaleway$0.003UnknownUnknown1
Voxtral Small 24B 2507Mistral$0.002$0.00232K4
Whisper Large v3OpenAI$0.002$0.002Unknown3
KB Whisperevroc$0.002$0.0024481
Voxtral Mini 3Bprivatemode-ai$0.005Unknown32K1
All-MiniLM-L6-v2digitalocean$0.009Unknown2561
Multi-QA-mpnet-base-dot-v1digitalocean$0.009Unknown5121
BGE Reranker v2 M3digitalocean$0.010Unknown8K1
Whisper large-v3privatemode-ai$0.016Unknown4481
Llama 3.2 1B InstructMeta$0.010$0.01060K7
Prompt Guard 2 86MMeta$0.010$0.0105122
Llama Prompt Guard 2 22MMeta$0.010$0.0105122
BGE M3digitalocean$0.020Unknown8K1
E5 Large v2digitalocean$0.020Unknown5121
Llama 3.2 3B InstructMeta$0.020$0.020131K10
PaddleOCR-VLnovita-ai$0.020$0.02016K1
Llama 3.1 8B InstructMeta$0.020$0.030131K18
Mistral Nemo Instruct 2407Mistral$0.020$0.030131K7
Nomic Embed Text v1.5tinfoil$0.050Unknown8K1
DeepSeek OCR 2DeepSeek$0.030$0.0304K3
Stable Diffusion 3.5 Largedigitalocean$0.080Unknown2561
Ministral 3Bazure-cognitive-services$0.040$0.040128K1
Ministral 3Bazure$0.040$0.040128K1
Ministral 3B (latest)Mistral$0.040$0.040128K1
Ling-3.0-flashopenrouter$0.021$0.063262K1
Llama 3 8B LunarisMeta$0.040$0.0508K2
GTE Large (v1.5)digitalocean$0.090Unknown8K1
Sao10k L3 8B Lunaris novita-ai$0.050$0.0508K1
L3 8B Stheno V3.2novita-ai$0.050$0.0508K1
Llama 3.2 11B Vision InstructMeta$0.055$0.055128K4
Qwen3.5 4BAlibaba (Qwen)$0.040$0.07033K3
MythoMax 13Bopenrouter$0.060$0.0608K1
Gemma 4 E4B ITGoogle$0.020$0.100131K3
Sarvam 30Bfastrouter$0.020$0.100128K1
Nex N2 Mininano-gpt$0.025$0.100262K1
Nex-N2-Miniopenrouter$0.025$0.100262K1
Granite 4.0 H Microcloudflare-workers-ai$0.017$0.112131K1
Granite 4.0 Microopenrouter$0.017$0.112131K1
Llama 3.1 8BMeta$0.050$0.080131K2
Sarvam 30Bnano-gpt$0.028$0.11166K1
baichuan-m2-32bnovita-ai$0.070$0.070131K1
LFM2-24B-A2Btogetherai$0.030$0.12033K1
LFM2 24B A2Bpioneer$0.030$0.12033K1
Granite 4.1 8Bnano-gpt$0.050$0.100131K1
Granite 4.1 8Bopenrouter$0.050$0.100131K1
Mellum2 12B A2.5Bwandb$0.050$0.100131K1
Granite 4.1 8Bwandb$0.050$0.100131K1
R1 Distill Llama 70BDeepSeek$0.030$0.1408K4
Gemma 3n 4BGoogle$0.060$0.12033K4
Phi 4 Multimodalnano-gpt$0.070$0.110128K1
Laguna XS 2.1openrouter$0.060$0.120262K1
Mythomax L2 13Bnovita-ai$0.090$0.0904K1
Command R7BCohere$0.037$0.150128K4
Command R7B ArabicCohere$0.037$0.150128K1
Qwen3.5 9BAlibaba (Qwen)$0.040$0.150262K22
Trinity Miniclarifai$0.045$0.150131K1
Trinity Minivercel$0.045$0.150131K1
Ministral 3 3B 2512Mistral$0.100$0.100131K3
Llama-3.2-3BMeta$0.100$0.100131K2
GPT OSS 120Bsynthetic$0.100$0.100131K1

전체 626개 중 상위 60개 표시. 추가 필터링은 전체 목록을 이용하세요.

Frequently asked questions

How many AI models support 오픈 웨이트?

626 canonical models in our database currently support 오픈 웨이트. The list is regenerated on every data refresh, so it always reflects the latest releases tracked in our catalogue.

What is the cheapest model with 오픈 웨이트?

Whisper Large v3 from scaleway is currently the lowest-priced option, at $0.003 per 1M input tokens and Unknown per 1M output tokens. The full table above is sorted price-ascending.

Which model with 오픈 웨이트 has the largest context window?

Llama 4 Scout 17B Instruct (US) (Meta) leads on context at 3.50M tokens. This may matter if you also need long-document understanding alongside 오픈 웨이트.

Which models are available on the most providers?

Production-readiness usually correlates with how many independent providers host the same weights. The top three by provider count are: GLM-5.2 (75), Kimi K2.6 (63), GLM-5.1 (54).

How is 오픈 웨이트 different from a regular LLM?

Open-weight models publish their trained weights publicly. You can self-host on your own GPUs, fine-tune on private data or run offline. Note: open weights ≠ open source — training data and code are usually not released.

How often is this list updated?

Daily. Our data pipeline syncs once a day, regenerates the canonical model list, and rebuilds these pages so newly released models appear within 24 hours.

마지막 업데이트:

Prices in USD per 1M tokens. Unknown means the provider does not publish per-token pricing.

Pricing and capabilities are refreshed daily and reconciled against each provider's official documentation. Always verify critical production decisions with the provider directly.