z-ai
Z.ai: GLM 4.7
z-ai/glm-4.7
Our take
AI-assisted analysis based on the sources below. Suitability is an editorial assessment; test it on your own workload.
GLM-4.7 (z-ai/glm-4.7) is Z.ai's text-only flagship model, distributed under an MIT license and listed in the OpenRouter catalog with a roughly 200K-token context window plus tool-calling and structured-output support. Vendor documentation positions it as a coding-focused release, reporting gains over GLM-4.6 in agentic programming, terminal tasks, tool use and reasoning, and describes interleaved, preserved and per-turn thinking modes. Most benchmark figures are supplier-reported, so independent validation on your own workload is advisable.[1][2]
Strengths
- Vendor-reported gains in agentic and multilingual coding and in terminal tasks versus GLM-4.6, framed by the supplier as the model's core focus.[2]
- Thinking controls: reasoning interleaved before each action, preservation of reasoning across turns for long agent runs, and per-turn toggling to trade latency against accuracy.[2]
- The OpenRouter catalog lists tool calling, structured outputs, reasoning enabled by default and a roughly 200K-token context window.[1]
Limitations
- Only text input and output are listed; no vision, audio or image capability is documented in this source pack.[1]
- Evidence is limited to the OpenRouter catalog and Z.ai's own model card; benchmark results are supplier-reported, and knowledge cutoff, safety behavior and real-world latency are unverified, so validate on your own workloads.[1][2]
- Local serving currently requires nightly or main-branch builds of vLLM or SGLang per the vendor documentation, which may raise operational effort.[2]
Where to use it
- Agentic coding and terminal workflows in frameworks such as Claude Code, Cline or Roo Code, the vendor's stated focus; pilot it before committing.[2]
- Long-horizon, multi-step tool-using agents where preserved cross-turn reasoning could improve consistency.[2]
- UI, webpage and slide generation, which the vendor calls vibe coding, plus general chat or creative writing per its documentation.[2]
Sources
- [1] OpenRouter catalog · Z.ai: GLM 4.7Sources checked: Nov 13, 2026
- [2] Model documentation · Z.ai: GLM 4.7Sources checked: Nov 13, 2026
About this model
Description below is the original text supplied by OpenRouter, shown in the source language.
GLM-4.7 is Z.ai’s latest flagship model, featuring upgrades in two key areas: enhanced programming capabilities and more stable multi-step reasoning/execution. It demonstrates significant improvements in executing complex agent tasks while...
Specifications
- Model ID
- z-ai/glm-4.7
- Canonical ID
- z-ai/glm-4.7-20251222
- Creator
- z-ai
- Added to OpenRouter
- 22.12.2025
- Context length
- 204,800
- Modality
- text->text
- Input modalities
- text
- Output modalities
- text
- Token pricing
- USD per 1M tokens: in $0.600 / out $2.2
- Knowledge cutoff
- —
- Expiration date
- —
Tasks and capabilities
A grade appears only where the source supplies a matching measurement for that task, the model is active and the cohort holds at least 10 comparable models. The capability column lists facts the source declares — supported formats, reasoning and tool parameters — which are not measured quality.
| Task | Grade | Source-declared capability |
|---|---|---|
| General quality | No measurement | reasoning: optional · parameters: include_reasoning, reasoning |
| Coding | BB — next 30% of the cohort | — |
| Agents | No measurement | parameters: structured_outputs, tool_choice, tools |
| Writing | No measurement | output: text |
| Reasoning | No measurement | reasoning: optional · parameters: include_reasoning, reasoning |
| Document analysis | No measurement | — |
| Vision | No measurement | — |
| Image generation | No measurement | — |
| Audio | No measurement | — |
| Video | No measurement | — |
| Embeddings | No measurement | — |
Evaluations
| Source | Benchmark | Value | Grade | Data retrieved |
|---|---|---|---|---|
| Artificial Analysis | Coding index | 45.3 | BB — next 30% of the cohort | 13.11.2026 |
| Design Arena | agents · androidnativewin rate 55.8% · rank 34 | 1,097Elo | CC — below the B band | 13.11.2026 |
| Design Arena | agents · fullstackwin rate 45.1% · rank 38 | 1,027Elo | CC — below the B band | 13.11.2026 |
| Design Arena | agents · godotgamedevwin rate 35.8% · rank 33 | 1,058Elo | CC — below the B band | 13.11.2026 |
| Design Arena | agents · mobileappswin rate 48.9% · rank 40 | 1,099Elo | CC — below the B band | 13.11.2026 |
| Design Arena | models · 3dwin rate 54.3% · rank 60 | 1,173Elo | BB — next 30% of the cohort | 13.11.2026 |
| Design Arena | models · asciiartwin rate 48.2% · rank 36 | 1,160Elo | BB — next 30% of the cohort | 13.11.2026 |
| Design Arena | models · codecategorieswin rate 54.8% · rank 61 | 1,207Elo | BB — next 30% of the cohort | 13.11.2026 |
| Design Arena | models · datavizwin rate 51.2% · rank 63 | 1,193Elo | BB — next 30% of the cohort | 13.11.2026 |
| Design Arena | models · gamedevwin rate 55.2% · rank 65 | 1,179Elo | BB — next 30% of the cohort | 13.11.2026 |
| Design Arena | models · svgwin rate 54.3% · rank 56 | 1,133Elo | CC — below the B band | 13.11.2026 |
| Design Arena | models · uicomponentwin rate 51.1% · rank 67 | 1,190Elo | BB — next 30% of the cohort | 13.11.2026 |
| Design Arena | models · websitewin rate 55.3% · rank 60 | 1,218Elo | BB — next 30% of the cohort | 13.11.2026 |
How grades work
Each letter grade compares a model only against other models scored by the same source and metric in the current snapshot. S means the top 10%, A the next 20%, B the next 30%, C the rest. Models without a published measurement are shown as not rated — a supported format alone never implies measured quality.
- S — S — top 10% of the cohort
- A — A — next 20% of the cohort
- B — B — next 30% of the cohort
- C — C — below the B band
Providers
Optional live provider statistics from OpenRouter. Missing values are unknown, not zero.
| Provider | Context length | Price in /1M | Price out /1M | Latency p50 (30m, s) | Throughput p50 (30m, tok/s) | Uptime (30m) |
|---|---|---|---|---|---|---|
| DeepInfrafp4 | 202,752 | $0.320 | $1.4 | — | — | — |
| DeepInfrafp4 | 202,752 | $0.400 | $1.75 | — | — | 100% |
| Venicefp4 | 198,000 | $0.400 | $1.93 | — | — | — |
| Novitafp8 | 204,800 | $0.600 | $2.2 | — | — | 100% |
| Z.AIfp4 | 202,752 | $0.600 | $2.2 | — | — | 100% |
| Mancer 2fp4 | 131,072 | $0.700 | $2.5 | — | — | 100% |
Provider data retrieved: 13.11.2026
Related articles
No published articles identify this model yet.