the AI bench
VERIFIED JUNE 2026
All models

MODEL · Z.AI (FORMERLY ZHIPU AI) · 744B TOTAL / 40B ACTIVE

GLM-5.1

Current #1 open-weight on SWE-Bench Pro (58.4) — a long-horizon agentic coding flagship that narrowly beats GPT-5.4 and Claude Opus 4.6 on that benchmark. MIT license means no commercial restrictions, unlike many frontier opens.

License: MIT · Context: 200K tokens (131K max output) · Released: April 7, 2026

The decision in five lines

The call
Hosted only
Best for
agents
Runs on
Hosted or workstation-class only · ~466 GB (not consumer-local)
Watch out
At Q4_K_XL it's 466 GB on disk and needs a data-center GPU partition or a Mac Studio 512 GB.
Evidence
Estimated · last verified April 2026

744B total
PARAMETERS
MOE
TYPE
200K
CONTEXT
~466 GB (not consumer-local)
VRAM AT Q4

Where we recommend this

Every tier slot in the planner where this model is a top or alternate pick. Pulled live from planner.js — when the planner refreshes, this table stays current.

AGENTS ·
GLM-5.1 (frontier, hosted)Long-horizon agentic specialist — stays productive across hundreds of rounds and thousands of tool calls. 744B MoE, MIT. Hosted via Z.ai or `glm-5.1:cloud` Ollama tag.
AGENTS · TOP
GLM-5.1 (frontier, hosted)744B MoE — SOTA on SWE-Bench Pro (58.4), 8-hour autonomous runs, MIT. Practical only via Z.ai hosted API or `glm-5.1:cloud` Ollama tag.

The call

Current #1 open-weight on SWE-Bench Pro (58.4) — a long-horizon agentic coding flagship that narrowly beats GPT-5.4 and Claude Opus 4.6 on that benchmark. MIT license means no commercial restrictions, unlike many frontier opens.

When not to use: Any single-card local setup. At Q4_K_XL it's 466 GB on disk and needs a data-center GPU partition or a Mac Studio 512 GB. For most users, the honest path is `glm-5.1:cloud` via Ollama rather than pretending it runs locally.

Runner notes

Unsloth dynamic 2-bit (220 GB) or 1-bit (200 GB) GGUFs make workstation-class local runs *possible* but slow. Ollama has `glm-5.1` + `glm-5.1:cloud` tags — `:cloud` is the honest choice for consumer hardware.

License
MIT
Released
April 7, 2026
Maker
Z.ai (formerly Zhipu AI)

Next step

Find-by-model — see what hardware runs this