Console/Models / clavue-2.1

Clavue 2.1

Flagship quality for hard reasoning and long agent loops.

premiumv2.1.0128K context−75%

11 days left

$0.435 / $0.87list / 1M$0.145 / $0.29sale · in/out

List prices match industry frontier API rates; sale price is what you pay now. Light (Fast) −70%, others −75%. Reverts after the promo window.

Overview

Clavue 2.1 is the flagship product model for hard reasoning, long agent loops, and multi-step tool work. In product role it sits with the class of assistants people reach for when Claude- or GLM-grade depth is expected — without juggling multiple vendor keys.

Model id: clavue-2.1
Released: 2026-07-01
Input: text
Output: text
Sale · input: $0.145 / 1M
Sale · output: $0.29 / 1M
List · input: $0.435 / 1M
List · output: $0.87 / 1M

List prices align with industry frontier API bands; launch sale is −70% (light) / −75% (others) vs list. All official models are available on every plan — only quota differs. Checkout: Alipay / WeChat today.

Global: Paid $9 · ×5 $19 · ×20 $39. API list matches industry frontier rates with a limited launch discount. Checkout today: Alipay / WeChat (CNY SKUs). Card billing soon.

Capability snapshot

Full table →

SWE-Bench Pro

57.8

Terminal-Bench v2.1

64.5

MCP-Atlas

70.5

IFBench

80.5

SysBench

94.2

MRCR-128k

90.2

Internal Clavue evaluation suite (2026-07). Methodology aligns with public agent benchmarks; not an independent third-party audit. Peer figures are approximate public references for context only.

Best for

Hard coding
Multi-agent
Deep analysis

Premium pool · flagship launch promo (−75% off list).

What builders say

★★★★★
“Switched our nightly agent loops to Clavue 2.1 Pro. Quality is in the same band as the big US chat models, and the bill finally looks rational.”
K. Nakamura · Staff engineer · multi-agent tooling
★★★★★
“If you’ve used Claude for planning and GLM for code, Clavue 2.1 sits in the middle with one bill and OpenAI-compatible wiring.”
Omar S. · Open-source maintainer
★★★★☆
“The launch promo is aggressive. List looks like frontier API rates; sale is what you’d hope a challenger charges.”
Lina V. · Agent platform PM
★★★★★
“I don’t need five vendors. Official models + membership quota is enough to feel like I’m underpaying for over-delivery.”
周琪 · Full-stack · Cursor → imux migrant

API

OpenAI-compatible · Base URL https://api.clavue.com

POST /v1/chat/completions

GET /v1/models

curl -s https://api.clavue.com/v1/chat/completions \
  -H "Authorization: Bearer $CLAVUE_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "clavue-2.1",
    "messages": [
      {"role": "user", "content": "Hello from Clavue"}
    ]
  }'

Changelog

v2.1.0 · 2026-07-01
Clavue 2.1 flagship product id; improved tool-loop reliability.
v2.0.0 · 2026-06-20
2.0 generation preview for internal agents.