Zhipu AI dropped GLM-5.2 on June 13 with full MIT-licensed open weights now live on Hugging Face, featuring a claimed usable 1M-token context and immediate availability through its GLM Coding Plan tiers for tasks that stretch across hours of agentic work.
BREAKING: GLM-5.2 is now 1st on Design Arena.
With an Elo of 1360, GLM-5.2 has jumped ahead of the now unavailable Claude Fable 5.
And it's open weights.
This is an improvement of 4 positions and 27 Elo points to achieve one of the highest Elo scores in our code categories since Design Arena started.
Huge congratulations to the @Zai_org on the release!
The 744B-parameter MoE model downloads freely and runs via vLLM or Transformers, while API pricing lands at $1.40 per million input tokens, positioning it as a cheaper alternative to closed coding suites.
Design Arena shows GLM-5.2 at 1360 Elo in code categories ahead of several Claude variants, yet independent third-party benchmarks remain absent at launch so the gap on long-horizon suites stays unverified outside company and arena data.
Many users praised GLM-5.2 topping the Design Arena Code Leaderboard at 1360 Elo as proof open-weights models can lead closed ones, while others called the results fake or useless in practice.
No Digg Deeper questions have been answered for this story yet.
Absolutely fucking crazy
BREAKING: GLM-5.2 is now 1st on Design Arena.
With an Elo of 1360, GLM-5.2 has jumped ahead of the now unavailable Claude Fable 5.
And it's open weights.
This is an improvement of 4 positions and 27 Elo points to achieve one of the highest Elo scores in our code categories since Design Arena started.
Huge congratulations to the @Zai_org on the release!
BREAKING: GLM-5.2 is now 1st on Design Arena.
With an Elo of 1360, GLM-5.2 has jumped ahead of the now unavailable Claude Fable 5.
And it's open weights.
This is an improvement of 4 positions and 27 Elo points to achieve one of the highest Elo scores in our code categories since Design Arena started.
Huge congratulations to the @Zai_org on the release!
It's hard to pinpoint open-closed gap and so-on, but I trust the @arena team and just look where GLM 5.2 is on this. An MIT licensed, to be open weight model. At this point you could argue they have a better agent than Gemini does. That's a serious accomplishment.
Still hard to expect the unexpected with AI. It goes to show how skilled many of the scientists are in China. They're hitting high peaks with much less compute.
Overall, I think the US models are really ahead, but you can't just discount the Chinese labs. Not at all.
BREAKING: GLM-5.2 is now 1st on Design Arena.
With an Elo of 1360, GLM-5.2 has jumped ahead of the now unavailable Claude Fable 5.
And it's open weights.
This is an improvement of 4 positions and 27 Elo points to achieve one of the highest Elo scores in our code categories since Design Arena started.
Huge congratulations to the @Zai_org on the release!
this is pretty insane and that's only 744B
imagine a properly RL'd DeepSeek-V4-Pro
BREAKING: GLM-5.2 is now 1st on Design Arena.
With an Elo of 1360, GLM-5.2 has jumped ahead of the now unavailable Claude Fable 5.
And it's open weights.
This is an improvement of 4 positions and 27 Elo points to achieve one of the highest Elo scores in our code categories since Design Arena started.
Huge congratulations to the @Zai_org on the release!
I don't trust arenas much. Waiting for AA scores I get the feeling this is the real one. Actual, solid, all-around excellent model, genuine frontier class
BREAKING: GLM-5.2 is now 1st on Design Arena.
With an Elo of 1360, GLM-5.2 has jumped ahead of the now unavailable Claude Fable 5.
And it's open weights.
This is an improvement of 4 positions and 27 Elo points to achieve one of the highest Elo scores in our code categories since Design Arena started.
Huge congratulations to the @Zai_org on the release!
Impressive if true!
Better than Claude Fable 5? Wow!
Design is really lacking in these frontier models, so I'm very curious to test GLM-5.2 myself.
Testing this already on a few internal use cases and will report back on findings.
BREAKING: GLM-5.2 is now 1st on Design Arena.
With an Elo of 1360, GLM-5.2 has jumped ahead of the now unavailable Claude Fable 5.
And it's open weights.
This is an improvement of 4 positions and 27 Elo points to achieve one of the highest Elo scores in our code categories since Design Arena started.
Huge congratulations to the @Zai_org on the release!
Looks Like GLM 5.2 was the real Le Chatton Fat
Absolutely fucking crazy
Looks strong at SWE too.
GLM 5.2 ranks #3 on FrontierSWE. It is only behind Fable 5 and Opus 4.8, and it outperforms GPT-5.5.
This is the first model that closes the large gap between models from Anthropic / OpenAI and other providers, and it is the strongest open-weight model by far.
GLM-5.2 (Max) by @Zai_org ranks #10 on the new Agent Arena leaderboard, closely matching Claude-Opus-4.8 (non-thinking) and is the #1 open model by a wide margin!
In Agent Arena, we measure models on millions of real-world, long-horizon agentic tasks from a global community of users. Models can access web search, filesystem, and terminal tools to complete complex workflows. The leaderboard measures model performance on outcomes relative to the average model using a causal tracing methodology.
Compared to 5.1, GLM-5.2 (Max) climbs from #13 to #10. Its clearest gains are confirmed task success, and user praise vs. complaint. Bash capabilities and tool hallucination remain stable. There is a tradeoff in steerability compared to the previous model (-6.0% vs. +1.2%).
GLM-5.2 remains the same price as GLM-5.1, $1.4/$4.4 per input/output MTokens. 1M context window.
Huge congrats @Zai_org for the incredible release!
See thread for details on how GLM-5.2 (Max) performs across 5 different signals.
Introducing GLM-5.2: Frontier Intelligence, Open Weights
- Significant improvements in coding and agentic tasks - Strong long-horizon capabilities with a 1M context window - Two levels of reasoning effort: GLM-5.2 (max) pushes the limits, while GLM-5.2 (high) strikes a strong balance between performance and token efficiency - MIT-licensed open weights - Same API pricing as GLM-5.1
Tech Blog: http://z.ai/blog/glm-5.2 Weights: http://huggingface.co/zai-org/GLM-5.2 API: http://docs.z.ai/guides/llm/glm-5.2 Coding Plan: http://z.ai/subscribe Chat: http://chat.z.ai
Exciting news: GLM-5.2 (Max) ranks #2 in Code Arena: Frontend, with +29pt over Claude Opus 4.7 (Thinking) and only behind Fable 5! GLM-5.2 is the best open model vs Kimi-K2.6 and Minimax-M3 by a large margin.
- #2 React and #4 HTML sub-leaderboards - Ranks as the top model in nearly all sub categories: Brand & Marketing, Reference-Based Design, Data & Analytics, Consumer Product, Gaming, and Simulations.
Congrats @Zai_org for the incredible milestone!
GLM-5.2 (Max) by @Zai_org ranks #10 on the new Agent Arena leaderboard, closely matching Claude-Opus-4.8 (non-thinking) and is the #1 open model by a wide margin!
In Agent Arena, we measure models on millions of real-world, long-horizon agentic tasks from a global community of users. Models can access web search, filesystem, and terminal tools to complete complex workflows. The leaderboard measures model performance on outcomes relative to the average model using a causal tracing methodology.
Compared to 5.1, GLM-5.2 (Max) climbs from #13 to #10. Its clearest gains are confirmed task success, and user praise vs. complaint. Bash capabilities and tool hallucination remain stable. There is a tradeoff in steerability compared to the previous model (-6.0% vs. +1.2%).
GLM-5.2 remains the same price as GLM-5.1, $1.4/$4.4 per input/output MTokens. 1M context window.
Huge congrats @Zai_org for the incredible release!
See thread for details on how GLM-5.2 (Max) performs across 5 different signals.
> sonnet lol
Every 3 weeks, some “open source” glup shitto model pops into existence from nowhere, scores a bunch of big numbers on the goybenches, and then everyone forgets them in a day because they’re always palpably worse than like, Claude Sonnet Medium. Who even cares?
Try it now on http://DesignArena.ai
Damn! That looks nice.
Exciting news: GLM-5.2 (Max) ranks #2 in Code Arena: Frontend, with +29pt over Claude Opus 4.7 (Thinking) and only behind Fable 5! GLM-5.2 is the best open model vs Kimi-K2.6 and Minimax-M3 by a large margin.
- #2 React and #4 HTML sub-leaderboards - Ranks as the top model in nearly all sub categories: Brand & Marketing, Reference-Based Design, Data & Analytics, Consumer Product, Gaming, and Simulations.
Congrats @Zai_org for the incredible milestone!
Just to be clear, if you remove Fable which is unavaialble, GLM-5.2 (Max) is the #1 model in the world for frontend coding.
This is a huge moment. OSS has caught up with proprietary, and China has caught up with the US, in this very important domain.
Exciting news: GLM-5.2 (Max) ranks #2 in Code Arena: Frontend, with +29pt over Claude Opus 4.7 (Thinking) and only behind Fable 5! GLM-5.2 is the best open model vs Kimi-K2.6 and Minimax-M3 by a large margin.
- #2 React and #4 HTML sub-leaderboards - Ranks as the top model in nearly all sub categories: Brand & Marketing, Reference-Based Design, Data & Analytics, Consumer Product, Gaming, and Simulations.
Congrats @Zai_org for the incredible milestone!
GLM-5.2 (Max) is #25 overall in the Text Arena, similar to GLM-5.1. Diving deeper into the comparison in Text, we can see that though GLM-5.2’s rank is down overall, its largest gains are across: - Sub-categories: Expert Arena, Multi-Turn - Occupational categories: such as: Life, Physical & Social Science, Creative Writing and Medicine & Healthcare.
GLM-5.2 ranks as the best open model in Code Arena: Frontend.
In the Code Arena: Frontend, models are evaluated on agentic frontend coding tasks from real users building apps and websites (HTML and React).
GLM-5.2 (Max) ranks #10 overall (+4.4%) - tied for #1 Tool Hallucination (+1.9%) - #3 Confirmed Task Success (+9.4%) - #3 Praise vs. Complaint (+14.9%) - #16 Bash Recovery (+1.7%) - #20 Steerability (-6.0%)
Learn more about the causal tracing methodology for Agent Arena on our blog: http://arena.ai/blog/agent-arena-methodology
@natolambert @arena it's about where you would expect it with a 6 month backward looking gap, still short of Opus 4.6
It's hard to pinpoint open-closed gap and so-on, but I trust the @arena team and just look where GLM 5.2 is on this. An MIT licensed, to be open weight model. At this point you could argue they have a better agent than Gemini does. That's a serious accomplishment.
