Model Center
The public BatchIn catalog prioritizes real live models, exclusives, and near-term launches. No filler, no fake cards.
38 models
Z.ai
glm-5.1
SWE-Bench Pro #1
Open-source coding flagship built for long-horizon autonomous engineering and deep reasoning.
Cached Input: $0.175
TTFT
520ms
Throughput
42 tok/s
Z.ai
glm-5
Lower-cost GLM route for production reasoning, agents, and long-context workflows.
Cached Input: $0.122
TTFT
520ms
Throughput
42 tok/s
Z.ai
glm-4.7
SWE-bench 73.8%
Mid-tier GLM reasoning route for engineering teams that need quality below flagship spend.
Cached Input: $0.052
TTFT
420ms
Throughput
58 tok/s
DeepSeek
deepseek-r1
o1-class reasoning
Heavy reasoning model for difficult planning, math, research, and multi-step analysis.
Cached Input: $0.063
TTFT
160ms
Throughput
120 tok/s
DeepSeek
deepseek-v3.2
IMO + IOI gold
Flagship DeepSeek release tuned for strong general reasoning at a very aggressive price point.
Cached Input: $0.035
TTFT
160ms
Throughput
120 tok/s
DeepSeek
deepseek-v3.1-terminus
Higher-output DeepSeek route for workflows that need longer structured completions.
Cached Input: $0.035
TTFT
160ms
Throughput
120 tok/s
DeepSeek
deepseek-v3
Stable general-purpose DeepSeek route for large-scale chat and batch workloads.
Cached Input: $0.028
TTFT
160ms
Throughput
120 tok/s
Alibaba
qwen3-32b
Balanced mid-large Qwen route for general chat, coding, and production assistant workloads.
Cached Input: $0.007
TTFT
220ms
Throughput
94 tok/s
Alibaba
qwen3.5-397b
201 languages
Top-tier Qwen MoE model for multilingual reasoning, coding, and large-context assistants.
Cached Input: $0.035
TTFT
420ms
Throughput
58 tok/s
Alibaba
qwen3.5-122b
Balanced Qwen MoE for long-context assistants and cost-conscious production routing.
Cached Input: $0.028
TTFT
310ms
Throughput
72 tok/s
Alibaba
qwen3.5-35b
Lower-cost MoE Qwen route for product copilots and high-volume assistant traffic.
Cached Input: $0.021
TTFT
220ms
Throughput
94 tok/s
Alibaba
qwen3.5-27b
exceeds GPT-5-mini
Lean Qwen route aimed at lower-cost chat, agent routing, and product copilot features.
Cached Input: $0.025
TTFT
220ms
Throughput
94 tok/s