Skip to main content

Model Library

Every model here trains on your agent's own data, benchmarks against your production model, and serves through one OpenAI-compatible API. The weights are yours.

Train a model
Qwen
Compact

Qwen 3 4B

4B params · 128K context

Cost to train
from $0.70
Cost to run (1M tokens)
from $4.00
Train now
Qwen
Compact

Qwen 3.5 4B

4B params · 256K context

Cost to train
from $0.70
Cost to run (1M tokens)
from $4.00
Train now
Meta
Compact

Llama 3.2 3B Instruct

3B params · 128K context

Cost to train
from $0.55
Cost to run (1M tokens)
from $4.00
Train now
Qwen
Compact

Qwen 3.5 2B

2B params · 256K context

Cost to train
from $0.55
Cost to run (1M tokens)
from $4.00
Train now
Cisco
Compact

Antares 1B (Security)

1.8B params · 128K context

Cost to train
from $0.55
Cost to run (1M tokens)
from $4.00
Train now
Qwen
Compact

Qwen 3 1.7B

1.7B params · 128K context

Cost to train
from $0.55
Cost to run (1M tokens)
from $4.00
Train now
Liquid AI
Compact

LFM2.5 1.2B Instruct

1.2B params · 32K context

Cost to train
from $0.50
Cost to run (1M tokens)
from $2.50
Train now
Meta
Compact

Llama 3.2 1B Instruct

1B params · 128K context

Cost to train
from $0.50
Cost to run (1M tokens)
from $2.50
Train now
Qwen
Compact

Qwen 3.5 0.8B

0.8B params · 256K context

Cost to train
from $0.50
Cost to run (1M tokens)
from $2.50
Train now
Qwen
Compact

Qwen 3 0.6B

0.6B params · 128K context

Cost to train
from $0.50
Cost to run (1M tokens)
from $2.50
Train now
Qwen
Compact

Qwen 2.5 0.5B Instruct

0.5B params · 32K context

Cost to train
from $0.50
Cost to run (1M tokens)
from $2.50
Train now
Liquid AI
Compact

LFM2.5 350M

350M params · 32K context

Cost to train
from $0.50
Cost to run (1M tokens)
from $2.50
Train now
Cisco
Compact

Antares 350M (Security)

350M params · 32K context

Cost to train
from $0.50
Cost to run (1M tokens)
from $2.50
Train now
Liquid AI
Compact

LFM2.5 230M

230M params · 32K context

Cost to train
from $0.50
Cost to run (1M tokens)
from $2.50
Train now
Qwen
Small

Qwen 3.5 9B

9B params · 256K context

Cost to train
from $1.00
Cost to run (1M tokens)
from $6.00
Train now
Qwen
Small

Qwen 3 8B

8B params · 128K context

Cost to train
from $1.00
Cost to run (1M tokens)
from $6.00
Train now
Meta
Small

Llama 3.1 8B Instruct

8B params · 128K context

Cost to train
from $1.00
Cost to run (1M tokens)
from $6.00
Train now
Google
Small

Gemma 4 E4B

E4B (~8B) params · 128K context

Cost to train
from $0.75
Cost to run (1M tokens)
from $4.00
Train now
Meta
Small

Llama 3.1 8B Instruct

8B params · 128K context

Cost to train
from $1.00
Cost to run (1M tokens)
from $6.00
Train now
Qwen
Small

Qwen 2.5 7B Instruct

7B params · 128K context

Cost to train
from $1.00
Cost to run (1M tokens)
from $6.00
Train now
Google
Small

Gemma 4 E2B

E2B (~5.1B) params · 128K context

Cost to train
from $0.55
Cost to run (1M tokens)
from $4.00
Train now
Qwen
Mid

Qwen 3 Coder 30B MoE

30B MoE (3B active) params · 256K context

Cost to train
from $2.00
Cost to run (1M tokens)
from $8.00
Train now
Qwen
Mid

Qwen 3.8 27B

27B params · 256K context

Cost to train
from $2.00
Cost to run (1M tokens)
from $15.00
Train now
Qwen
Mid

Qwen 3.6 27B

27B params · 256K context

Cost to train
from $2.00
Cost to run (1M tokens)
from $15.00
Train now
Qwen
Mid

Qwen 3.5 27B

27B params · 256K context

Cost to train
from $2.00
Cost to run (1M tokens)
from $15.00
Train now
Open AI
Mid

GPT-OSS 20B

20B MoE params · 128K context

Cost to train
from $1.75
Cost to run (1M tokens)
from $12.00
Train now
Qwen
Mid

Qwen 3 14B

14B params · 128K context

Cost to train
from $1.25
Cost to run (1M tokens)
from $10.00
Train now
Qwen
Mid

Qwen 2.5 14B Instruct

14B params · 128K context

Cost to train
from $1.25
Cost to run (1M tokens)
from $10.00
Train now
Google
Mid

Gemma 4 12B

12B params · 256K context

Cost to train
from $1.00
Cost to run (1M tokens)
from $6.00
Train now
Qwen
Large

Qwen 2.5 72B Instruct

72B params · 128K context

Cost to train
from $5.00
Cost to run (1M tokens)
from $40.00
Train now
Meta
Large

Llama 3.3 70B Instruct

70B params · 128K context

Cost to train
from $4.50
Cost to run (1M tokens)
from $35.00
Train now
Qwen
Large

Qwen 3.5 35B MoE

35B MoE (3B active) params · 256K context

Cost to train
from $2.50
Cost to run (1M tokens)
from $8.00
Train now
Qwen
Large

Qwen 3 32B

32B params · 128K context

Cost to train
from $2.25
Cost to run (1M tokens)
from $20.00
Train now
Qwen
Large

Qwen 2.5 32B Instruct

32B params · 128K context

Cost to train
from $2.25
Cost to run (1M tokens)
from $20.00
Train now
Qwen
Large

Qwen 2.5 Coder 32B Instruct

32B params · 128K context

Cost to train
from $2.25
Cost to run (1M tokens)
from $20.00
Train now
Google
Large

Gemma 4 31B

31B params · 256K context

Cost to train
from $2.50
Cost to run (1M tokens)
from $10.00
Train now
NVIDIA
Large

Nemotron 3.5 Lightning 30B-A3B

30B-A3B params · 256K context

Cost to train
from $2.00
Cost to run (1M tokens)
from $6.00
Train now
Meta
Large

Muse Glimmer 30B

30B params · 128K context

Cost to train
from $2.50
Cost to run (1M tokens)
from $10.00
Train now
Google
Large

Gemma 4 26B-A4B

26B (4B active) params · 256K context

Cost to train
from $2.00
Cost to run (1M tokens)
from $8.00
Train now