GLM
Z.ai's open-weight family, behind the cheap coding subscriptions.
GLM is the model family from Z.ai, and the reason its coding plan turns up in every "cheaper than Claude Code" thread — the weights are published, plenty of providers serve them, and the Flash models are some of the cheapest usable coding models anywhere. The 5.3 line runs to a million tokens of context, with Flash variants that also read images and video. Z.ai's own chat app is listed separately under Chats.
At a glance — GLM 5.3
- Cost
- $1.40 in · $4.40 out per 1M tokensMid-range
- Memory
- 1.3M tokens — about 1,950 pages
- Reads
- Text
- Can do
- Use tools (agents)
- Structured output
- Thinking mode
- Good at
- Coding: better than 87% · Agentic work: better than 95% · Overall: better than 86%
- Reliability
- 99.98% uptime, last 24h
- Released
- Aug 18, 2026
Advanced details
- API model ID
z-ai/glm-5.3- Max output
- 944K tokens
- Cached input
- $0.26 per 1M
- Tokenizer
- Other
- Coding index
- 74.8
- Agentic work index
- 53.1
- Overall index
- 44.8
Providers
The same model, hosted by different companies. OpenRouter picks one per request and falls back to the next if it fails.
| Provider | Input /1M | Output /1M | Cached /1M | Max output | Uptime 24h |
|---|---|---|---|---|---|
| Baidu · fp8 | $0.561 | $1.764 | $0.104 | 131K | 99.98% |
| DeepInfra · fp4 | $0.563 | $2.50 | $0.125 | 131K | 98.65% |
| Morph | $0.583 | $1.833 | $0.096 | 944K | 99.76% |
| Inceptron · fp4 | $0.597 | $2.502 | $0.152 | 944K | 99.29% |
| InferenceNet | $0.684 | $2.28 | $0.114 | 131K | 99.74% |
| Relace | $0.70 | $2.20 | $0.13 | 131K | — |
| Reka | $0.761 | $2.574 | $0.152 | 236K | 99.81% |
| Io Net · fp8 | $0.77 | $2.62 | $0.14 | 66K | 99.74% |
| Sail Research · us | $0.77 | $4.00 | $0.19 | 944K | 99.93% |
| Sail Research · fp8 | $0.77 | $4.00 | $0.19 | 944K | 99.84% |
| Novita · fp8 | $0.783 | $2.46 | $0.145 | 131K | 99.85% |
| Phala | $0.84 | $2.64 | $0.156 | 131K | 99.50% |
| DigitalOcean | $0.91 | $2.86 | $0.169 | 128K | 99.27% |
| GMICloud · fp8 | $0.98 | $3.08 | $0.182 | 944K | 99.73% |
| Makora · fp4 | $1.05 | $4.20 | $0.19 | 128K | 98.69% |
| SiliconFlow · fp8 | $1.12 | $3.52 | $0.208 | 262K | 99.40% |
| Alibaba | $1.19 | $3.74 | $0.238 | 131K | 99.82% |
| Decart · fp4 | $1.19 | $3.74 | $0.196 | 944K | 99.76% |
| Friendli | $1.26 | $3.96 | $0.234 | 944K | 99.98% |
| AkashML · fp8 | $1.30 | $4.40 | $0.26 | 131K | 99.73% |
| Wafer · us | $1.40 | $4.40 | $0.26 | 944K | 99.86% |
| BaseTen · fp4Degraded | $1.40 | $4.40 | $0.14 | 262K | 99.27% |
| Mistral · nvfp4 | $1.40 | $4.40 | $0.14 | 131K | 99.91% |
| Crusoe · fp4 | $1.40 | $4.40 | $0.26 | 944K | 99.65% |
| PrimeIntellect | $1.40 | $4.40 | $0.26 | 131K | 99.97% |
| Wafer | $1.40 | $4.40 | $0.26 | 944K | 99.87% |
| Venice | $1.40 | $4.40 | $0.26 | 131K | 98.67% |
| Together | $1.40 | $4.40 | $0.26 | 944K | 99.83% |
| Parasail · fp8 | $1.40 | $4.40 | $0.26 | 944K | 99.57% |
| Modal | $1.40 | $4.40 | $0.26 | 944K | 99.38% |
| BaseTen · fp4Degraded | $1.40 | $4.40 | $0.14 | 262K | 98.06% |
| Fireworks | $1.40 | $4.40 | $0.26 | 944K | 98.81% |
| Cloudflare | $1.40 | $4.40 | $0.26 | 1.2M | 99.96% |
| AtlasCloud · fp8 | $1.40 | $4.40 | $0.26 | 131K | 99.70% |
| Z.AI · fp8 | $1.40 | $4.40 | $0.26 | 131K | 99.93% |
| BaseTen · fast | $2.10 | $6.60 | $0.21 | 262K | 99.71% |
| BaseTen · fast | $2.10 | $6.60 | $0.21 | 262K | 99.86% |
| Alibaba · fast | $2.80 | $8.80 | $0.56 | 131K | 99.70% |
Supported parameters
- frequency_penalty
- include_reasoning
- logit_bias
- logprobs
- max_tokens
- min_p
- parallel_tool_calls
- presence_penalty
- reasoning
- reasoning_effort
- repetition_penalty
- response_format
- seed
- stop
- structured_outputs
- temperature
- tool_choice
- tools
- top_k
- top_logprobs
- top_p
Live from OpenRouter, refreshed hourly. Scores from Artificial Analysis.
Key info
- Pricing
- Open source
- Category
- Models
- Best for
- Intermediate
- Made by
- Z.ai
- Weights
- Open
- Good for
- Coding on a budget