Z.ai · Released Sep 30, 2025
GLM 4.6
Compared with GLM-4.5, this generation brings several key improvements: Longer context window: The context window has been expanded from 128K to 200K tokens, enabling the model to handle more complex...
- Released
- Sep 30 2025
- Context window
- 205K 131K out
- Input / 1M tokens
- $0.55 from Free
- Output / 1M tokens
- $2.20 from Free
Where to run it
31 providers compared
Free to try on iFlow, ModelScope — free endpoints are often limited-time and rate-limited. Cheapest: AIHubMix at $0.27 / $1.10 per 1M input / output tokens — 61% less than the priciest option.
| Provider | Input | Output | Cache read | Context | Max output | Uptime | vs. best |
|---|---|---|---|---|---|---|---|
| iFlowfree · may be limited-time | Free | Free | — | 200K | 128K | — | Free |
| ModelScopefree · may be limited-time | Free | Free | — | 203K | 98K | — | Free |
| AIHubMix | $0.27 | $1.10 | $0.055 | 200K | 131K | — | Cheapest |
| 302.AI | $0.29 | $1.14 | — | 205K | 131K | — | +4% |
| ZenMux | $0.29 | $1.17 | $0.058 | 200K | 128K | — | +6% |
| DeepInfraFP4 | $0.40 | $1.60 | $0.08 | 203K | 131K | — | +46% |
| Helicone | $0.45 | $1.50 | — | 205K | 131K | — | +49% |
| VeniceFP4 | $0.43 | $1.75 | $0.08 | 198K | 16K | 84%24h | +58% |
| Venice AI | $0.43 | $1.75 | $0.08 | 198K | 16K | — | +58% |
| Meganova | $0.45 | $1.90 | — | 203K | 131K | — | +69% |
| DeepInfraFP4 | $0.50 | $2.00 | $0.10 | 203K | 131K | 99.6%24h | +82% |
| NanoGPT | $0.50 | $2.00 | $0.10 | 256K | 66K | — | +82% |
| IO.NET | $0.52 | $2.04 | $0.26 | 200K | 4K | — | +88% |
| NovitaBF16 | $0.55 | $2.20 | $0.11 | 205K | 131K | 99.8%24h | +101% |
| Kilo Gateway | $0.55 | $2.20 | $0.11 | 205K | 131K | — | +101% |
| LLM Gateway | $0.55 | $2.20 | $0.11 | 205K | 131K | — | +101% |
| Novita AI | $0.55 | $2.20 | $0.11 | 205K | 131K | — | +101% |
| Abacus | $0.55 | $2.20 | — | 203K | 131K | — | +101% |
| DevPass (LLM Gateway) | $0.55 | $2.20 | $0.11 | 205K | 131K | — | +101% |
| Hugging Face | $0.55 | $2.20 | — | 205K | 131K | — | +101% |
| Z.AIFP4 | $0.60 | $2.20 | $0.11 | 203K | 131K | 99.9%24h | +109% |
| Zhipu AI | $0.60 | $2.20 | $0.11 | 205K | 131K | — | +109% |
| Vercel AI Gateway | $0.60 | $2.20 | $0.11 | 200K | 96K | — | +109% |
| Tempr Gateway | $0.60 | $2.20 | $0.11 | 205K | 131K | — | +109% |
| Merge Gateway | $0.60 | $2.20 | $0.11 | 200K | 131K | — | +109% |
| Ofox | $0.60 | $2.20 | $0.11 | 205K | 131K | — | +109% |
| Impossibl | $0.60 | $2.20 | $0.11 | 205K | 131K | — | +109% |
| OrcaRouter | $0.60 | $2.20 | $0.11 | 205K | 131K | — | +109% |
| OpenCode | $0.60 | $2.20 | $0.10 | 205K | 131K | — | +109% |
| Eden AI | $0.60 | $2.20 | $0.11 | 203K | 131K | — | +109% |
| Poe | $0.61 | $2.22 | $0.11 | — | — | — | +111% |
| Qiniu | $1.04 | $1.83 | — | 200K | 200K | — | +158% |
Value
Price vs. performance
GLM 4.6 Other Z.ai Other labs Value frontier · Terminal-Bench vs. input price per 1M tokens (log scale) · Drag to pan, pinch or Ctrl + scroll to zoom, double-click to zoom in.
Benchmarks
4 benchmark scores
| Benchmark | Score | Rank | Leader |
|---|---|---|---|
| Terminal-BenchSoftware engineering · Tasks completed with the programs in a real terminal (best agent) | 24.5% | #35of 40 | GPT-5.584.7% |
| SciCodeSoftware engineering · Research-coding problems from the natural sciences | 38.4% | #82of 102 | Claude Opus 5.566.9% |
| WebDev ArenaDesign & visual · People vote on which model builds the better website (Elo) | 1341 | #77of 90 | Claude Opus 5.51820 |
| CritPtReasoning & knowledge · Open-ended, research-level physics problems | 1.1% | #76of 111 | GPT-5.6 Sol32.3% |
Specs
Details
- Model ID
z-ai/glm-4.6- Developer
- Z.ai
- Release date
- Sep 30, 2025 ·
- Context window
- 204,800 tokens
- Max output
- 131,072 tokens
- API price
- $0.55 / $2.20 per 1M input / output tokens
- Providers
- 31
- Input types
- text
- Reasoning
- Yes
- Tool calling
- Yes
- Open weights
- Yes
- Free tier
- No
Release date and prices from OpenRouter's catalog; providers from OpenRouter and models.dev. Buying directly from a provider can cost a little differently. Speed, latency and uptime are measured on OpenRouter only; uptime is the share of OpenRouter requests each provider served over the last day. Benchmark scores are the best result per benchmark from Epoch AI, licensed CC BY 4.0. Rank counts every model tracked here with a score.