About
Ring-2.6-1T is a 1T-parameter-scale thinking model with 63B active parameters, built for real-world agent workflows that require both strong capability and operational efficiency. It is optimized for coding agents, tool...
Pricing
Input
$0.07
per 1M tokens
Output
$0.63
per 1M tokens
Blended
$0.21
per 1M tokens
Cheaper than 71% of models. Median price is $0.67/1M tokens.
Cost Calculator
Tokens per day1M
100K100M
Daily
$0.21
Monthly
$6.38
vs. Similar Models
Grok 4.1 Fast (Reasoning)Q:0.0
$0.28+29%
OpenAI: GPT-5.1-Codex-MiniQ:0.0
$0.69+224%
Qwen3.5 Omni PlusQ:0.0
$1.50+606%
Qwen3.6 27B (Non-reasoning)Q:-0.1
$1.35+535%
Performance
124
tokens/sec
Faster than 66% of models
1.81
seconds
Faster than 28% of models
17.87
seconds
Faster than 28% of models
Market Median
93 tok/s
34% faster
Median TTFT
1.13s
60% slower
Throughput/Dollar
586
tok/s per $/1M
Speed Comparison
Qwen: Qwen3 VL 30B A3B Instruct
124 tok/s-0%
Qwen3 VL 30B A3B (Reasoning)
125 tok/s+1%
GPT-5.6 Terra (medium)
124 tok/s-1%
Context Window
262K
tokens
Larger than 60% of models
Max Output
66K
tokens
25% of context
Benchmarks
MMLU-ProNot evaluated
GPQA Diamond
85.7%
HLE
18.3%
LiveCodeBenchNot evaluated
SciCode
42.4%
TerminalBench Hard
28.8%
MATH-500Not evaluated
AIMENot evaluated
AIME 2025Not evaluated
IFBench
44.6%
Long Context Recall
64.3%
Tau2
92.4%
Market AverageTop Score