Native multimodal GLM model for efficient coding and long-horizon agent tasks
GLM-5.3-Flash by zhipuai is offered by 98 providers on this page. Public prices are shown for 86 of them. Its official list price is $0.15 per 1M input tokens and $0.50 per 1M output tokens. The lowest paid channel is Pareto Inference at $0.03 / $0.10 per 1M, about 15× below the list price. That channel is a third-party gateway, so confirm its availability and rate limits before depending on it. It also has 6 free ($0) channels and 6 covered by a paid subscription; free tiers usually carry rate limits, and subscription-covered access bills $0 per token only after the subscription fee.
Artificial Analysis rates it 41.8 on the Intelligence Index, with 71.5 for coding and 50.9 for agentic tasks. Against its lowest blended price of $0.048 per 1M, that is roughly 880 index points per dollar, which is the value ratio the leaderboards rank on. Median output speed is 60 tokens per second, with 3.02s to the first token. Running one task of the Artificial Analysis suite costs about $0.2533, which reflects how many tokens its reasoning consumes rather than the unit price alone.
The context window is 1,000,000 tokens at the reference host, but hosts report different limits, from 200,000 to 1,049,000, so the usable window depends on the provider you pick. It supports reasoning, tool use, and structured output. Accepted input modalities are Text, Image, Video, and PDF. The weights are open, so it can also be self-hosted or served through a gateway of your choice. Providers report the capability flags inconsistently, so verify a specific feature against the host you plan to use.
Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.
| Provider | Tier | Input | Output | Cache read | Cache write | Context | Output limit | Status |
|---|---|---|---|---|---|---|---|---|
| NaN glm5.3-flash | Gateway | Free | — | — | 1,000,000 | 131,072 | ||
| Umans AI Coding Plan umans-glm-5.3-flash | Gateway | Subscription | $0 | $0 | 1,048,576 ⚠ | 131,071 | ||
| Umans AI Coding Plan umans-coder | Gateway | Subscription | $0 | $0 | 1,048,576 ⚠ | 131,071 | ||
| Zhipu AI Coding Plan | Gateway | Subscription | $0 | $0 | 1,000,000 | 131,072 | ||
| Volcengine Ark Coding Plan | Gateway | Subscription | $0 | — | 1,000,000 | 131,072 | ||
| Z.AI Coding Plan | Gateway | Subscription | $0 | $0 | 1,000,000 | 131,072 | ||
| OrcaRouter z-ai/glm-5.3-flash-free | Gateway | Free | — | — | 1,000,000 | 128,000 | ||
| SCNet Token Plan GLM-5.3-Flash | Gateway | Subscription | $0 | — | 1,000,000 | 131,072 | ||
| AIHubMix coding-glm-5.3-flash-free | Gateway | Free | — | — | 1,000,000 | 131,072 | ||
| AIHubMix ox-alpha | Gateway | Free | — | — | 1,000,000 | 131,072 | ||
| Z.AIOfficial | First-party | $0.15 | $0.50 | $0.03 | $0 | 1,000,000 | 131,072 | |
Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.
Interleaved thinking (reasoning between tool calls) is declared by 51 of 98 providers.
4 / 98 providers expose no reasoning control (reasoning_options: []).
| Mode | Provider | Input | Output | Cache read | Cache write |
|---|---|---|---|---|---|
| priority | Fireworks AI | $0.19 | $0.63 | $0.038 | — |
| priority | Fireworks AI | $0.19 | $0.63 | $0.038 | — |
12 more channels offer $0 (NaN, Umans AI Coding Plan, Umans AI Coding Plan etc.); free tiers usually have rate limits and no SLA, excluded from ranking.
| Name | Conditions | Score | Metric | Source |
|---|---|---|---|---|
| Terminal-Bench | harness: Claude Code 2.1.207 · variant: max effort · v2.1 | 84.3 | pass@1 | Source ↗ |
| DeepSWE | harness: mini-swe-agent · variant: max effort · v1.1 | 63.4 | resolved | Source ↗ |
| Agents' Last Exam | variant: max effort | 26.3 | score | Source ↗ |
| AutomationBench | variant: max effort · v1.0.6 | 48.8 | pass@1 | Source ↗ |
| Humanity's Last Exam | variant: max effort; with tools · dataset: full set | 55.3 | accuracy | Source ↗ |
| GDPval-AA | variant: max effort · v2 | 1773 | Elo | Source ↗ |
The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.
Individual evaluations run by Artificial Analysis, on the same reasoning tier as the intelligence score above. Each benchmark has its own task set and harness, so rows are not comparable with one another. The Intelligence Index above draws on a different, newer set of evaluations.