GLM-5.3
- Next shutdown
- None setNo date on 4 platforms
- Arena Text
- 1,478#17 of 128 · Arena
- GPQA Diamond
- 90.9%#27 of 108 · Epoch AI
- Context window
- 1Mtokens
- Released
- Aug 18, 2026Z.ai (GLM)
Dates on every platform
| Platform | Model ID and status | Deprecated | Shuts down | Replacement |
|---|---|---|---|---|
| Amazon BedrockAmazon Web Services: Amazon Bedrock ListFoundationModels, us-east-1 (opens in a new tab) · checked Oct 14, 2026 | zai.glm-5.3Active · also global.zai.glm-5.3, us.zai.glm-5.3 | – | None set | – |
| Mistral APIMistral AI: Z.ai GLM 5.3 model card (opens in a new tab) · checked Oct 14, 2026 | zai-glm-5-3Active · also zai-glm-5, zai-glm-latest | – | None set | – |
| Together AITogether AI: Changelog (opens in a new tab) · checked Oct 14, 2026 | zai-org/GLM-5.3Active | – | None set | – |
| Fireworks AIFireworks AI: GLM-5.3 model page (opens in a new tab) · checked Oct 14, 2026 | accounts/fireworks/models/glm-5p3Active | – | None set | – |
Bedrock commits to at least 45 days of notice before end of life for this model.
In us-east-1, requests go through a cross-Region inference profile (listed in aliases).
Third-party model from Z.ai hosted by Mistral; third-party models get one month of deprecation notice.
The zai-glm-5 alias points to GLM 5.3, not to GLM 5.
What breaks when you move off GLM-5.3
No breaking changes are recorded for GLM-5.3.
Intelligence
| Benchmark | Publisher | Score | Rank | Published |
|---|---|---|---|---|
| Arena Text | Arena | 1,478 ± 3 | #17 of 128 | Oct 8, 2026 |
| GPQA Diamond | Epoch AI | 90.9% ± 1.6% | #27 of 108 | Aug 24, 2026 |
| Arena WebDev | Arena | 1,622 ± 4 | #12 of 69 | Oct 8, 2026 |
| Arena Agent | Arena | 0 ± 0 | #22 of 35 | Oct 12, 2026 |
| FrontierMath Tier 4 | Epoch AI | 29.3% ± 7.2% | #29 of 58 | Aug 25, 2026 |
| FrontierMath Tiers 1-3 | Epoch AI | 68.8% ± 2.7% | #24 of 71 | Aug 25, 2026 |
| OTIS Mock AIME 2024-2025 | Epoch AI | 91.1% ± 3.5% | #44 of 105 | Aug 24, 2026 |
| SimpleQA Verified | Epoch AI | 41.0% ± 1.6% | #39 of 68 | Aug 28, 2026 |
Arena (arena.ai), https://arena.ai/leaderboard
Epoch AI, ‘Capabilities & benchmarking’. Published online at epoch.ai. Retrieved from ‘https://epoch.ai/benchmarks’ [online resource].
List prices
| Platform | Input | Cached | Output | In effect | Source |
|---|---|---|---|---|---|
Mistral APIzai-glm-5-3 | $1.40 | $0.14 | $4.40 | Sep 15, 2026Current | Mistral AI: API pricing (opens in a new tab) |
Together AIzai-org/GLM-5.3 | $1.40 | $0.26 | $4.40 | Aug 28, 2026Current | Together AI: Available models (serverless) (opens in a new tab) |
Fireworks AIaccounts/fireworks/models/glm-5p3 | $1.40 | $0.26 | $4.40 | Oct 14, 2026Current · Standard serverless mode. Batch inference is billed at half the Standard rate; Priority, Fast and US-only endpoints cost more. | Fireworks AI: Serverless Pricing (opens in a new tab) |
Change history
ShutdownZ.ai GLM 5.2 retires from the Mistral APIMistral API
zai-glm-5-2 retires on 31 October 2026. Mistral names Z.ai GLM 5.3 (zai-glm-5-3) as the replacement at the same price.
Source: Mistral AI, Changelog (opens in a new tab); Mistral AI, Models Overview (opens in a new tab)
ReleaseGLM 5.3 available on Amazon BedrockAmazon Bedrock
Bedrock added GLM 5.3 as zai.glm-5.3, with at least 45 days of notice before any end of life.
- Platform
- Amazon Bedrock
- Model
- GLM-5.3
Source: Amazon Web Services, Amazon Bedrock ListFoundationModels, us-east-1 (opens in a new tab)
ReleaseZ.ai GLM 5.3 generally available on the Mistral APIMistral API
zai-glm-5-3 reached General Availability on the Mistral API. The zai-glm-5 and zai-glm-latest aliases point to it.
- Platform
- Mistral API
- Model
- GLM-5.3
Source: Mistral AI, Changelog (opens in a new tab); Mistral AI, Z.ai GLM 5.3 model card (opens in a new tab)
ReleaseGLM-5.3 arrives on Together serverlessTogether AI
zai-org/GLM-5.3 is available on serverless with a context length of one million tokens, function calling and structured outputs.
- Platform
- Together AI
- Model
- GLM-5.3
Source: Together AI, Changelog (opens in a new tab); Together AI, Available models (serverless) (opens in a new tab)