vs
Z.ai vs Boundless
Z.ai builds GLM and serves GLM-5.3 at $1.40 in, $4.40 out from servers mostly in China. Boundless serves GLM-5.3 at $1.12 in, $3.52 out in the US or Europe.
By The Subconscious Team · Updated
Z.ai vs Boundless: key differences
Z.ai serves its own GLM-5.3 at $1.40 in and $4.40 out per million tokens with a 1M context, cached input at $0.26 and about 80 tokens per second. GLM-5.3-Flash costs $0.075 in and $0.25 out, older Flash models are free, and the GLM Coding Plan starts at $18 a month. Its servers sit mostly in China. Boundless serves GLM-5.3 at $1.12 in and $3.52 out and GLM-5.3 Flash at $0.12 in and $0.40 out, both with 1M context, and Vercel measured them at 134 and 276 tokens per second. Requests run in the US or Europe with zero data retention.
Z.ai is the pick for the cheapest Flash tokens, free older models and a monthly Coding Plan that works inside Claude Code. Boundless fits teams that want GLM-5.3 for less per token, served outside China, or a reserved deployment that its engineers tune against contract targets. Both offer OpenAI and Anthropic compatible APIs.
What Z.ai and Boundless do
Z.ai
Z.AI is the international brand of Chinese lab Zhipu AI, maker of the GLM models. Its current flagship, GLM-5.3, shipped August 17, 2026 at $1.40 in and $4.40 out per million tokens, with cached input at $0.26. Its weights came out August 28 under a custom GLM-5.3 License that requires a security review for hosts with more than $10B in yearly revenue. GLM-5.3-Flash costs $0.075 in and $0.25 out, and several older Flash models are priced at zero, a real free tier instead of trial credits. GLM-5, released in February 2026, is a 744B mixture-of-experts model under an MIT license, and at launch it ranked first among open-weight models on the Artificial Analysis index with a record-low hallucination score.
Example models: GLM-5.3, GLM-5.3-Flash
Full Z.ai profileBoundless
Boundless launched in September 2026 as an inference partner for AI-native startups. It sells two things. The Model API serves pre-tuned models on shared capacity, billed per token from prepaid credits with a budget per API key. Tailored inference reserves capacity for one customer, and Boundless engineers pick the model on the customer’s own evals, tune precision, batching, caching, speculative decoding and parallelism, then run the deployment against uptime, throughput and latency targets written into the contract. A free two-week evaluation tests those targets on real traffic against the customer’s current provider first.
Example models: GLM-5.3, DeepSeek V4.1 Flash, Kimi K3
Full Boundless profileShould you choose Z.ai or Boundless?
Z.ai vs Boundless at a glance
| Attribute | Z.ai | Boundless |
|---|---|---|
| Model access | Open weights (GLM-5.3 License; MIT on Flash and older) | Open weights, plus closed GPT and Claude |
| Flagship models | GLM-5.3, GLM-5.3-Flash | GLM-5.3, DeepSeek V4.1 Flash, Kimi K3 |
| Speed | ~80 tok/s on GLM-5.3 | 342 tok/s on DeepSeek V4.1 Flash 1017 (Vercel) |
| Price | $1.40 in, $4.40 out (GLM-5.3); free Flash tier | $0.04–$2.30 in, $0.21–$11.40 out per 1M (open) |
| Customization | Open weights; GLM-5.3 License adds review for large hosts | Serving-stack tuning, no fine-tuning |
| Deployment | API, GLM Coding Plan | Serverless API, reserved tailored capacity |
| Long context | 1M (GLM-5.3) | 1M context on most models |
Frequently asked questions
What is the difference between Z.ai and Boundless?
Z.ai builds GLM and serves GLM-5.3 at $1.40 in, $4.40 out from servers mostly in China. Boundless serves GLM-5.3 at $1.12 in, $3.52 out in the US or Europe.
When should I choose Z.ai over Boundless?
GLM Coding Plan from $18 a month; Free older Flash models; Cheapest GLM-5.3-Flash tokens.
When should I choose Boundless over Z.ai?
Lower GLM-5.3 prices; GLM served in the US or Europe; Engineer-tuned reserved capacity.
Is Z.ai or Boundless cheaper?
Z.ai: $1.40 in, $4.40 out (GLM-5.3); free Flash tier. Boundless: $0.04–$2.30 in, $0.21–$11.40 out per 1M (open). The cheaper choice depends on the model and workload.
Which has more context, Z.ai or Boundless?
Z.ai: 1M (GLM-5.3). Boundless: 1M context on most models.
Related comparisons
Running long-horizon agents?
If your agents run past 200K tokens, compare both against Subconscious. Our inference stack treats a long-horizon trace as the primary workload, so speed, cost, and accuracy hold up deep into the trace.