All models and templates

Robot policies you run as your own deployment, on GPUs close to your robots: start from one of these — seq init pi05-droid, then seq deploy pi05-droid/policy.py — or bring a checkpoint you trained.

policy

π0.5 DROID

pi05-droid

7 joint velocities and the gripper at 15 Hz · L4

$seq init pi05-droid

policy

π0.5 LIBERO

pi05-libero

π0.5 for the LIBERO suites · Franka Panda

$seq init pi05-libero

policy

GR00T N1.7 DROID

gr00t-droid

7 absolute joint positions and the gripper at 15 Hz · L4

$seq init gr00t-droid

policy

Cosmos 3 Edge DROID

cosmos-droid

7 absolute joint positions and the gripper at 15 Hz, from three cameras · L4

$seq init cosmos-droid

policy

π0.5 DROID + Qwen planner

pi05-droid-qwen

Qwen3.8 27B names the next step every ~2 s; π0.5 runs it at 15 Hz

$seq init pi05-droid-qwen

policy

π0.5 RoboCasa

pi05-robocasa

12 end-effector and base actions at 20 Hz for RoboCasa's PandaOmron · L4

$seq init pi05-robocasa

policy

π0.5 RoboTwin

pi05-robotwin

14 joint targets for two Aloha-AgileX arms at 30 Hz, 50 at a time · L4

$seq init pi05-robotwin

The simulations a policy is scored on, each in the environment it will run in: start from one — seq init libero, then seq deploy libero/benchmark.py. Every one is on the Benchmarks page, grouped by the arm layout it scores.

benchmark

LIBERO

libero

40 tasks · Franka Panda · four suites — spatial, object, goal and long-horizon · language-conditioned

$seq init libero

benchmark

RoboCasa365

robocasa

18 atomic tasks · PandaOmron, a Franka arm on a mobile base · RoboCasa365's pretraining kitchens

$seq init robocasa

benchmark

RoboTwin 2.0

robotwin

50 dual-arm tasks · Aloha-AgileX · the clean setting: no clutter, a plain table, fixed light

$seq init robotwin

The reasoning half: messages in, a message out, with its usage and cost on every answer.

vlm

Claude Opus 5.5

claude-opus-5-5

$4.2 / $21 per Mtok · images in · claude-opus-5-5

vlm

Claude Fable 5.1

claude-fable-5-1

$10.5 / $52.5 per Mtok · images in · claude-fable-5-1

vlm

GPT-6 Sol

gpt-6-sol

$2.1 / $10.5 per Mtok · images in · gpt-6-sol

vlm

GPT-6 Astra

gpt-6-astra

$10.5 / $52.5 per Mtok · images in · gpt-6-astra

vlm

Gemini 3.1 Pro (preview)

gemini-3-1-pro

$2.1 / $12.6 per Mtok · images in · gemini-3-1-pro

vlm

Gemini 3.8 Flash

gemini-3-8-flash

$0.7875 / $3.9375 per Mtok · images in · gemini-3-8-flash

vlm

Qwen3.8 27B

qwen3-8-27b

$0.44625 / $2.6775 per Mtok · images in · qwen3-8-27b

vlm

Perceptron Mk1.5

perceptron-mk1-5

$0.1575 / $1.575 per Mtok · images in · perceptron-mk1-5

vlm

Claude Haiku 5.5

claude-haiku-5-5

$0.105 / $0.525 per Mtok · images in · claude-haiku-5-5

vlm

GPT-6 Luna

gpt-6-luna

$0.105 / $0.525 per Mtok · images in · gpt-6-luna

vlm

GLM 5.3 Flash

glm-5-3-flash

$0.1575 / $0.525 per Mtok · images in · glm-5-3-flash

Not a chat model — it answers typed questions about a state (a yes/no probability, a pick from your options, or a score on a rubric) and returns calibrated probabilities for your own code to act on. No image input, no generated text.

decision

Jev

jev

$0.0441 / M in · structured out · jev