Chinese foundation model developer Z.ai (Zhipu AI) has opened public API access to GLM-5.3, offering developers direct endpoint integration following the model's initial release. The company kept base token rates aligned with the prior generation while introducing discounted prompt caching.
GLM-5.3 is priced at $1.40 per million input tokens and $4.40 per million output tokens on the Z.ai platform. For workloads utilizing prompt caching, cached input tokens are billed at $0.26 per million, an 81.4% discount relative to standard inputs. Z.ai noted that cache storage fees are currently waived during the initial rollout phase.
API Architecture and Protocol Support
Access is structured around standard OpenAI Chat Completions protocol conventions, allowing developers to route prompts to glm-5.3 with drop-in client compatibility. Developers previously holding GLM Coding Plan subscriptions can access the endpoint immediately.
At $5.80 for a combined one million input and one million output token benchmark profile, GLM-5.3 sits well below Western proprietary frontier endpoints:
- GLM-5.3: $1.40 input / $4.40 output ($5.80 / 1M pair)
- Grok 4.6 (standard context): $2.00 input / $6.00 output ($8.00 / 1M pair)
- Kimi K3: $3.00 input / $15.00 output ($18.00 / 1M pair)
- Claude Opus 5: $5.00 input / $25.00 output ($30.00 / 1M pair)
- GPT-5.6 Sol (standard): $5.00 input / $30.00 output ($35.00 / 1M pair)
Lightweight flash-tier alternatives remain cheaper, including Gemini 3.7 Flash at $0.75/$3.75 and GPT-5.6 Luna at $0.20/$1.20, but target different capability tiers.
Benchmark Positioning vs. Real-World Task Economics
Independent evaluations from Artificial Analysis place GLM-5.3 at an Intelligence Index score of 60 when evaluated in maximum reasoning mode. This score matches Moonshot AI's Kimi K3 as the highest-ranking open-weights architecture, sitting seven points above GLM-5.2 (53).
However, benchmark tracking highlights an important operational caveat in inference economics. While token rates remain flat between GLM-5.2 and GLM-5.3, the estimated cost per Artificial Analysis benchmark task rose from $0.44 on GLM-5.2 to $0.68 on GLM-5.3. The 54.5% increase in task completion expense stems from the model's deeper reasoning traces and expanded chain-of-thought generation before returning final answers.
Z.ai reiterated that full open-weight model checkpoints will be released publicly, though exact distribution dates and license terms have not been finalized.
