14 models
Ranked by Terminal-Bench Hard
Reasoning Tools JSON Open weights 35.6

Newer StepFun flash model for faster agents, coding, and multimodal prompts

stepfun/step-3.7-flash 2026-05-29 256K context $0.202/M input $1.211/M output
21 providers
Reasoning Tools JSON Open weights 32.6

StepFun flash model for efficient multimodal reasoning, coding, and tool use

stepfun/step-3.5-flash-2603 2026-04-02 256K context $0.105/M input $0.314/M output
6 providers
Reasoning Tools JSON Open weights 27.3

StepFun flash lane for quick multimodal reasoning and coding assistance

stepfun/step-3.5-flash 2026-01-29 256K context $0.105/M input $0.314/M output
15 providers
Reasoning Tools Open weights 25.0

Late GLM-4 workhorse for coding agents, reasoning, and structured tasks

zhipuai/glm-4.6 2025-09-30 204.8K context $0.6/M input $2.2/M output
21 providers
Reasoning Tools Open weights 22.0

Hybrid-reasoning GLM release that made the 4.5 line broadly useful

zhipuai/glm-4.5 2025-07-28 131.072K context $0.6/M input $2.2/M output
15 providers
Reasoning Tools Open weights 20.5

Lighter GLM-4.5 variant for fast coding assistance and cheaper agents

zhipuai/glm-4.5-air 2025-07-28 131.072K context $0.12/M input $0.896/M output
13 providers
Tools Open weights 18.9

Mistral's coding-agent model for repository work, terminal tasks, and software fixes

mistral/devstral-2512 2025-12-09 262.144K context $0.4/M input $2/M output
13 providers
Reasoning Tools Open weights 17.4

Fast Mistral production model for chat, extraction, and cost-sensitive agents

mistral/mistral-small-2603 2026-03-16 256K context $0.15/M input $0.6/M output
17 providers
Tools Open weights 15.9

Mistral's largest general model for enterprise agents, coding, and multilingual reasoning

mistral/mistral-large-2512 2025-12-02 262.144K context $0.5/M input $1.5/M output
16 providers
Tools Open weights 15.2

Smaller Qwen coder for efficient local agents and repo-level fixes

alibaba/qwen3-coder-30b-a3b-instruct 2025-07-31 262.144K context $0.45/M input $2.25/M output
13 providers
Reasoning Open weights 6.1

Classic open reasoning model for transparent math, coding, and deliberate problem solving

deepseek/deepseek-r1 2025-01-20 128K context $0.58/M input $2.319/M output
16 providers
Open weights 6.1

Flagship Mistral model for advanced reasoning, coding, and multilingual work

mistral/mistral-large-2411 2024-11-18 131.072K context Input not listed Output not listed
2 providers
Reasoning Tools Open weights 5.3

GLM vision model for visual reasoning, documents, and multimodal agents

zhipuai/glm-4.5v 2025-08-11 64K context $0.299/M input $0.896/M output
10 providers
Tools Open weights 3.0

Popular open Llama workhorse for multilingual chat, coding, and self-hosting

meta/llama-3.3-70b-instruct 2024-12-06 128K context $0.1/M input $0.32/M output
29 providers