Model library

Every base model on the platform, with the context window it takes, the highest LoRA rank it accepts, and what it costs to fine-tune and to serve. Prices are USD per million tokens.

49 models

Language

NVIDIA

Nemotron 3.5 Lightning

nvidia/Nemotron-3.5-Lightning

MoE · 12B params · 1.5B active

Context 128K LoRA rank 64

$0.22/MFine-tune

$0.15/MServe

OpenAI

GPT-OSS 20B

openai/gpt-oss-20b

MoE · 21B params · 3.6B active

Context 128K LoRA rank 64

$0.36/MFine-tune

$0.25/MServe

NVIDIA

Nemotron 3 Nano 30B

nvidia/Nemotron-3-Nano-30B-A3B

MoE · 30B params · 3B active

Context 64K LoRA rank 64

$0.40/MFine-tune

$0.27/MServe

Qwen

Qwen3 8B

Qwen/Qwen3-8B

Dense · 8B params

Context 32K LoRA rank 64

$0.40/MFine-tune

$0.30/MServe

Qwen

Qwen3.8 27B

Qwen/Qwen3.8-27B-A3B

MoE · 27B params · 3B active

Context 256K LoRA rank 64

$0.44/MFine-tune

$0.31/MServe

Qwen

Qwen3.6 35B

Qwen/Qwen3.6-35B-A3B

MoE · 35B params · 3B active

Context 256K LoRA rank 64

$0.48/MFine-tune

$0.34/MServe

Z.ai

GLM 5.3 Flash

zai-org/GLM-5.3-Flash

MoE · 106B params · 12B active

Context 1M LoRA rank 32

$0.60/MFine-tune

$0.40/MServe

OpenAI

GPT-OSS 120B

openai/gpt-oss-120b

MoE · 117B params · 5.1B active

Context 128K LoRA rank 32

$0.68/MFine-tune

$0.45/MServe

MiniMax

MiniMax M3

MiniMaxAI/MiniMax-M3

MoE · 456B params · 46B active

Context 1M LoRA rank 16

$0.72/MFine-tune

$0.48/MServe

Mistral AI

Mistral Medium 3.5

mistralai/Mistral-Medium-3.5

Dense · 41B params

Context 256K LoRA rank 32

$1.05/MFine-tune

$0.70/MServe

DeepSeek

DeepSeek V4 Flash 0731

deepseek-ai/DeepSeek-V4-Flash-0731

MoE · 284B params · 13B active

Context 1M LoRA rank 32

$1.70/MFine-tune

$1.10/MServe

DeepSeek

DeepSeek V4 Flash

deepseek-ai/DeepSeek-V4-Flash

MoE · 284B params · 13B active

Preview, superseded by 0731 · serve only · withdrawn 31 Jan 2027

Context 1MServe only

n/aFine-tune

$1.10/MServe

Llama 3.3 70B

meta-llama/Llama-3.3-70B

Dense · 70B params

Context 128K LoRA rank 32

$2.75/MFine-tune

$1.90/MServe

DeepSeek

DeepSeek V3.2

deepseek-ai/DeepSeek-V3.2

MoE · 685B params · 37B active

Context 128K LoRA rank 16

$3.20/MFine-tune

$2.10/MServe

DeepSeek

DeepSeek V3.1

deepseek-ai/DeepSeek-V3.1

MoE · 671B params · 37B active

Withdrawn 31 Jan 2027

Context 128K LoRA rank 16

$3.40/MFine-tune

$2.30/MServe

Z.ai

GLM 5.2

zai-org/GLM-5.2

MoE · 753B params · 40B active

Context 1M LoRA rank 16

$3.95/MFine-tune

$2.55/MServe

Moonshot AI

Kimi K2.6

moonshotai/Kimi-K2.6

MoE · 1T params · 32B active

Withdrawn 31 Jan 2027

Context 128K LoRA rank 16

$4.40/MFine-tune

$2.75/MServe

Z.ai

GLM 5.3

zai-org/GLM-5.3

MoE · 780B params · 40B active

Context 1M LoRA rank 16

$4.60/MFine-tune

$2.95/MServe

NVIDIA

Nemotron 3 Ultra 550B

nvidia/Nemotron-3-Ultra-550B-A55B

MoE · 550B params · 55B active

Context 64K LoRA rank 16

$5.00/MFine-tune

$3.10/MServe

DeepSeek

DeepSeek V4 Pro

deepseek-ai/DeepSeek-V4-Pro

MoE · 1.6T params · 49B active

Context 1M LoRA rank 16

$5.60/MFine-tune

$3.50/MServe

Moonshot AI

Kimi K3

moonshotai/Kimi-K3

MoE · 2.8T params · 104B active

Context 1MServe only

n/aFine-tune

$6.50/MServe

Vision

Google

Gemma 4 E4B

google/gemma-4-E4B

Dense · 8B params · 4.5B effective

Context 128K LoRA rank 64

$0.35/MFine-tune

$0.24/MServe

Qwen

Qwen3-VL 30B Instruct

Qwen/Qwen3-VL-30B-A3B-Instruct

MoE · 30B params · 3B active

Context 128K LoRA rank 64

$0.45/MFine-tune

$0.32/MServe

Qwen

Qwen3-VL 8B Instruct

Qwen/Qwen3-VL-8B-Instruct

Dense · 8B params

Context 128K LoRA rank 64

$0.48/MFine-tune

$0.35/MServe

Qwen

Qwen3.5 9B

Qwen/Qwen3.5-9B

Dense · 9B params

Context 64K LoRA rank 64

$1.30/MFine-tune

$0.90/MServe

Google

Gemma 4 31B

google/gemma-4-31B

Dense · 31B params

Context 256K LoRA rank 32

$1.45/MFine-tune

$0.95/MServe

Qwen

Qwen3-VL 235B Instruct

Qwen/Qwen3-VL-235B-A22B-Instruct

MoE · 235B params · 22B active

Context 128K LoRA rank 32

$2.40/MFine-tune

$1.60/MServe

Qwen

Qwen3.5 397B

Qwen/Qwen3.5-397B-A17B

MoE · 397B params · 17B active

Context 64K LoRA rank 32

$6.00/MFine-tune

$3.75/MServe

Image generation

Black Forest Labs

FLUX.2 klein

black-forest-labs/FLUX.2-klein

Flow transformer · 4B params

Context Up to 2MP LoRA rank 64

$0.50/MFine-tune

$2.50/MServe

Stability AI

Stable Diffusion 3.5 Large

stabilityai/stable-diffusion-3.5-large

MMDiT · 8B params

Context Up to 2MP LoRA rank 64

$0.90/MFine-tune

$4.50/MServe

Qwen

Qwen-Image

Qwen/Qwen-Image

MMDiT · 20B params

Context Up to 4MP LoRA rank 32

$1.30/MFine-tune

$7.50/MServe

Black Forest Labs

FLUX.2 dev

black-forest-labs/FLUX.2-dev

Flow transformer · 32B params

Context Up to 4MP LoRA rank 32

$1.40/MFine-tune

$8.00/MServe

Video generation

Tencent

HunyuanVideo 1.5

tencent/HunyuanVideo-1.5

DiT · 8.3B params

Context 720p · 10s LoRA rank 64

$1.40/MFine-tune

$5.00/MServe

Lightricks

LTX-2

Lightricks/LTX-2

DiT · native audio + video

Context Up to 4K · 10s LoRA rank 32

$1.60/MFine-tune

$5.50/MServe

Wan AI

Wan 2.2 T2V

Wan-AI/Wan2.2-T2V-A14B

MoE DiT · 27B params · 14B active

Context 720p · 5s LoRA rank 32

$1.80/MFine-tune

$6.50/MServe

Audio

OpenAI

Whisper large-v3 turbo

openai/whisper-large-v3-turbo

ASR · 809M params

Context 30s windows LoRA rank 64

$0.28/MFine-tune

$0.18/MServe

NVIDIA

Parakeet TDT 1.1B

nvidia/parakeet-tdt-1.1b

ASR · 1.1B params

Context Streaming LoRA rank 64

$0.30/MFine-tune

$0.20/MServe

NVIDIA

Sortformer Diarizer

nvidia/diar_sortformer_4spk-v1

Diarisation · 115M params

Context Up to 4 speakers LoRA rank 64

$0.35/MFine-tune

$0.24/MServe

NVIDIA

Nemotron 3.5 ASR

nvidia/Nemotron-3.5-ASR-Multilingual

ASR · 1.1B params · 25 languages

Context Streaming LoRA rank 64

$0.45/MFine-tune

$0.30/MServe

Qwen

Qwen3 TTS 1.7B

Qwen/Qwen3-TTS-1.7B

TTS · 1.7B params · streaming

Context 10 languages LoRA rank 64

$0.50/MFine-tune

$2.50/MServe

OpenAI

Whisper large-v3

openai/whisper-large-v3

ASR · 1.5B params

Context 30s windows LoRA rank 64

$0.60/MFine-tune

$0.40/MServe

Canopy Labs

Orpheus 3B

canopylabs/orpheus-3b-0.1-ft

TTS · 3B params · streaming

Context 8 voices LoRA rank 64

$0.70/MFine-tune

$3.50/MServe

Mistral AI

Voxtral 4B TTS

mistralai/Voxtral-4B-TTS-2603

TTS · 4B params

Context 9 languages LoRA rank 64

$0.90/MFine-tune

$4.50/MServe

Embedding & reranking

Google

EmbeddingGemma 300M

google/embeddinggemma-300m

Bi-encoder · 300M params · 768 dims

Context 2K LoRA rank 64

$0.03/MFine-tune

$0.02/MServe

BAAI

BGE Reranker v2 M3

BAAI/bge-reranker-v2-m3

Cross-encoder · 568M params · 100+ languages

Context 8K LoRA rank 64

$0.04/MFine-tune

$0.03/MServe

BAAI

BGE EN ICL

BAAI/bge-en-icl

Bi-encoder · 7B params · 4096 dims

Context 32K LoRA rank 64

$0.08/MFine-tune

$0.05/MServe

Qwen

Qwen3 Embedding 8B

Qwen/Qwen3-Embedding-8B

Bi-encoder · 8B params · 4096 dims

Context 32K LoRA rank 64

$0.08/MFine-tune

$0.05/MServe

Nomic AI

Nomic Embed Code

nomic-ai/nomic-embed-code

Bi-encoder · 7B params · 3584 dims

Context 32K LoRA rank 64

$0.08/MFine-tune

$0.05/MServe

Qwen

Qwen3 Reranker 8B

Qwen/Qwen3-Reranker-8B

Cross-encoder · 8B params

Context 32K LoRA rank 64

$0.09/MFine-tune

$0.06/MServe