Skip to content
ModelCap

Inference provider

Phala: AI models and API pricing

Every tracked model Phala serves through OpenRouter, with Phala's own input and output price per 1M tokens, the context window it serves, disclosed quantization and measured endpoint uptime. Rows link to the model page, where every provider's deployment is listed side by side.

Snapshot as of 1 October 2026

Phala serves 20 tracked models through 20 endpoints as of 1 October 2026; the highest-ranked is Kimi K3 (ModelCap Index #10); mean measured uptime is 97.62%.

Models served
20
On the ModelCap Index
12
Cheapest output price
$0.20/1M

Language models on Phala (20)

Ranked models first, in ModelCap Index order. Prices are the lowest Phala lists for the model; the endpoint count shows how many deployments carry it.

Language models served by Phala, with price, context, quantization and uptime
#ModelModelCap IndexInput $/1MOutput $/1MContextQuantizationUptime (30 min)Endpoints
1Kimi K3Moonshot AI#1082.2$2.55$12.751M—99.61%1
2GLM 5.3Z.ai#1180.9$0.84$2.641M—99.86%1
3DeepSeek V4.1 FlashDeepSeek#1579.3$0.21$0.841M—99.89%1Open weights
4GLM 5.3 FlashZ.ai#1778.9$0.12$0.401Mfp899.50%1Open weights
5DeepSeek V4 Pro 0813DeepSeek#2476.4$0.957$2.881M—100.0%1Open weights
6Muse Glimmer 30BMeta#3470.7$0.30$1.10131K——1Open weights
7Qwen3.8 27BQwen#3968.9$0.15$1.881M—99.17%1Open weights
8Qwen3.5 397B A17BQwen#4368.0$0.55$3.50262K—99.86%1Open weights
9DeepSeek V3.2DeepSeek#6661.7$1.00$1.00164K—99.62%1Open weights
10Qwen3.6 35B A3BQwen#7260.7$0.20$1.27262K—99.89%1Open weights
11Nemotron 3.5 LightningNVIDIA#8756.3$0.07$0.20262K—100.0%1
12gpt-oss-120bOpenAI#13244.6$0.15$0.60131K—100.0%1Open weights
13Qwen2.5 7B InstructQwen—$0.10$0.2033K—100.0%1Open weights
14DeepSeek V4 Flash 0731DeepSeek—$0.308$0.9241M—100.0%1Open weights
15GLM 5.1Z.ai—$1.21$4.20203K——1Open weights
16GLM 5.2Z.ai—$1.26$3.001Mfp8100.0%1Open weights
17Hy3Tencent—$0.15$0.64262K—100.0%1Open weights
18Kimi K2.6Moonshot AI—$1.09$4.60262K—100.0%1
19Qwen3.5-27BQwen—$0.30$2.40262K—63.59%1Open weights
20Qwen3.6 27BQwen—$0.32$2.70262K—96.14%1Open weights

Phala on ModelCap: common questions

How many AI models does Phala serve?

20 tracked models as of 1 October 2026: 20 language models (12 with a public ModelCap Index position), across 20 live endpoints observed through OpenRouter.

What is the best model available on Phala?

Kimi K3 from Moonshot AI is the highest-ranked language model Phala serves, at ModelCap Index position #10 and $12.75 per 1M output tokens on Phala. The table lists every model in Index order.

What is the cheapest model on Phala?

Nemotron 3.5 Lightning at $0.20 per 1M output tokens ($0.07 per 1M input), the lowest positive language-token price Phala lists. Free-tier endpoints, where present, show as $0 and are excluded from that comparison.

Which Phala model has the largest context window?

Kimi K3, served with a 1M-token context window on Phala. Providers sometimes serve less than the model's published maximum, so the figure here is the provider's own.

How reliable is Phala?

Mean measured uptime across Phala endpoints is 97.62% over the last 30 minutes, as reported by OpenRouter at the time of the snapshot. Per-model uptime is in the table; it is a live operational signal, not a service-level guarantee.

Where do the Phala prices come from?

From the OpenRouter catalogue entry for each Phala endpoint, in USD per 1M tokens, re-observed on every ModelCap refresh. Where Phala exposes several deployments of one model at different prices, the table shows the lowest and notes the count; the model page lists each deployment.

How current is this page?

This page uses the published ModelCap snapshot dated 1 October 2026. Publication and page updates happen separately. Offers retained from a failed refresh are marked stale in the data and are not counted here.

Explore further