讯飞星火

maas-api.cn-huabei-1.xf-yun.com

iFlytek Spark MaaS platform offering Spark series LLMs with strong Chinese language capabilities via OpenAI-compatible API.

Models: 10 models
From: --
Speed: 41 tok/s
Updated: 7/8/2026

Latency: 0.00 s
Created At: 8/13/2025

API Endpoints

Endpoint 1
maas-api.cn-huabei-1.xf-yun.com
Domain registered 12/30/2013
Endpoint 2
spark-api-open.xf-yun.com
Endpoint 3
maas-coding-api.cn-huabei-1.xf-yun.com

Claim this provider

Verify ownership to unlock provider management features:

Edit provider name, content, and links
Get featured with priority traffic and visibility boost
Display a verified badge to build user trust

Leaderboard Rankings

Health

1.0%#158/716

Data as of Jul 8, 2026, 03:11 PMRankings are based on community-submitted tests and periodic health probes. Advisory only, not official data.

About 讯飞星火

讯飞星火 (iFlytek Spark) is a large language model providing API access for AI capabilities. It uses a Mixture of Experts (MoE) architecture for improved inference efficiency. Key features include:

Primary capabilities: Language understanding, knowledge Q&A, code generation, logical reasoning, mathematical problem-solving, and content creation.
Multimodal support: Includes text, speech recognition (37 languages), and speech synthesis with customizable parameters.
API performance: Stream-based design with millisecond-level first-frame response for fast processing.
Use cases: Enterprise services (e.g., document generation, meeting summaries), smart hardware, government affairs, finance, healthcare, and daily applications like AI search and virtual assistants.
Additional tools: Offers SparkDesk desktop and mobile apps, AI search with deep web sourcing, and agent creation for customized solutions.
Security: Cloud services meet Level 3 cybersecurity standards with data encryption and access controls.

Health Check

100%Recent availability

History (72 pts)

PastNow

API Benchmarks & Pricing

Compare 10 model rows across audit recency, latest speed tests, throughput, latency, and per-token pricing.

Model	Speed	Latency
xopkimik26	38.2 t/s	1.40 s
xopqwen36v35b	-	-
xopglm51	22.2 t/s	7.49 s
xop35qwen2b	-	-
astron-code-latest	47.7 t/s	9.54 s
xopdeepseekv32	33.7 t/s	1.22 s
4.0Ultra	23.2 t/s	1.62 s
lite	53.8 t/s	1.53 s
xdeepseekr1	13.9 t/s	62.16 s
xop3qwen1b7	52.0 t/s	0.97 s

Showing 10 of 10 model rows

Recent Test Records

Time	Model	Speed	Latency
Jul 29, 03:16 AM	astron-code-latest	17.97 tok/s	11.79s
Jul 29, 03:04 AM	astron-code-latest	19.73 tok/s	5.78s
Jul 25, 11:33 AM	xopkimik26	32.24 tok/s	1.37s
Jul 15, 07:10 AM	xopkimik26	44.20 tok/s	1.43s
Jul 15, 06:47 AM	xopglm51	21.59 tok/s	6.80s
Jul 9, 03:02 PM	astron-code-latest	37.53 tok/s	1.60s
Jul 2, 12:57 AM	astron-code-latest	188.25 tok/s	6.01s
Jun 27, 10:56 PM	xopglm51	26.24 tok/s	3.02s
Jun 18, 08:33 AM	xopglm51	18.64 tok/s	12.63s
May 31, 01:44 AM	astron-code-latest	82.11 tok/s	3.90s

Similar API Provider Alternatives to Compare

Compare 讯飞星火 alternatives against 6 nearby API providers using 61 LMSpeed signals across shared model coverage, pricing, benchmark speed, uptime, and free-model availability.

Provider	Why compare	Models	Avg price	Speed	30d uptime
讯飞星火 iflytek-spark iFlytek Spark MaaS platform offering Spark series LLMs with strong Chinese language capabilities via OpenAI-compatible API.	Current provider baseline	10	N/A	41 tok/s	9970%
Mistral AI mistral-ai-api Mistral AI API provides access to frontier AI models including Mistral Large, Medium, Small, and Codestral for chat, code, and embeddings.	Faster measured speed Higher 30-day availability Broader model coverage Same provider category	11	N/A	132 tok/s	9980%
阿里云百炼 DashScope dashscope Alibaba Cloud DashScope provides AI model APIs including Qwen LLMs, vision, audio, and embedding models.	Faster measured speed Broader model coverage Same provider category	290	N/A	76 tok/s	7830%
百度千帆 baidu-qianfan Baidu Qianfan is Baidu Intelligent Clouds enterprise AI platform, providing access to ERNIE series models and other LLM capabilities through developer APIs.	Faster measured speed Broader model coverage Same provider category	184	N/A	81 tok/s	5020%
火山引擎 Ark volcengine-ark Volcengine Ark is ByteDance's enterprise AI platform, offering Doubao series models and third-party LLMs via OpenAI-compatible API.	Faster measured speed Broader model coverage Same provider category	163	N/A	41 tok/s	9960%
GLM BigModel Relay glm-bigmodel-relay A third-party API relay providing access to GLM/ChatGLM models through OpenAI-compatible endpoints.	Faster measured speed Broader model coverage Same provider category	140	$0.0083/M	52 tok/s	9970%
智谱 AI zhipu-ai Zhipu AI provides the GLM series of large language models including GLM-4, ChatGLM, and CodeGeeX for text, code, and multimodal tasks.	Faster measured speed Broader model coverage Same provider category	33	N/A	63 tok/s	9970%

Notes

Health checks: Scope: the 72-hour chart and recent availability measure API connectivity only. Each bar summarizes one hour of checks. Targets: LMSpeed tries the configured health check URL and provider status URL first, then API endpoints derived from known API hosts and recent speed-test base URLs. A website host is considered only when it looks like an API endpoint. Probe steps: each candidate goes through DNS lookup, TCP connection, TLS handshake for HTTPS, and an HTTP HEAD request with redirects followed. Probing stops after the first reachable candidate. Reachable criteria: every required network step must succeed. An HTTP response below 500 is treated as reachable, including 401 because it confirms that an authenticated API endpoint responded, except for statuses classified as blocked. Blocked results: HTTP 403, 429, 521, 525, and 530, plus detected WAF or Cloudflare challenges, are shown as blocked and excluded from availability calculations because LMSpeed cannot determine whether the API itself is down. Model availability: when a dedicated test key is configured, LMSpeed sends an authenticated GET request to a derived /models endpoint and compares returned model IDs with this provider's listed models. These per-model results appear in Models & Pricing and are not included in the provider connectivity percentage. Timeouts: TCP connection, TLS handshake, HTTP connectivity, and model requests each use a 20-second timeout. A full run can take longer when several candidates are tried. Frequency: a background worker checks all providers every 5 minutes by default. The 72-hour chart combines those samples into hourly bars, and the schedule may be changed by the service operator. Limit: automated samples are not an SLA and do not guarantee account quota, every model, every region, or successful completion requests. Check the provider's own status page before making operational decisions.
Domain Rating data is sourced from Ahrefs. It is a 0–100 backlink-based domain strength signal and does not measure API speed or reliability.
Announcements and FAQ are read from this provider's NewAPI status snapshot when available. LMSpeed stores the original content and optional English translations from the provider status source, then shows the localized fields on this page.