The web, organized for agents
Exa raises $250M Series CExa raises $250M Series C to build the search engine for AIs.Read more
smbtech.au/news/google-deepmind-releases-gemma-4-its-most-capable-open-source-ai-models
github.com/HiDream-ai/HiDream-O1-Image
bentoml.com/blog/navigating-the-world-of-open-source-large-language-models
smbtech.au/news/google-deepmind-releases-gemma-4-its-most-capable-open-source-ai-models
github.com/HiDream-ai/HiDream-O1-Image
bentoml.com/blog/navigating-the-world-of-open-source-large-language-models
smbtech.au/news/google-deepmind-releases-gemma-4-its-most-capable-open-source-ai-models
github.com/HiDream-ai/HiDream-O1-Image
bentoml.com/blog/navigating-the-world-of-open-source-large-language-models
smbtech.au/news/google-deepmind-releases-gemma-4-its-most-capable-open-source-ai-models
github.com/HiDream-ai/HiDream-O1-Image
bentoml.com/blog/navigating-the-world-of-open-source-large-language-models
smbtech.au/news/google-deepmind-releases-gemma-4-its-most-capable-open-source-ai-models
github.com/HiDream-ai/HiDream-O1-Image
fireworks.ai/blog/best-open-source-llms
github.com/MoonshotAI/Kimi-K2.5?tab=readme-ov-file
github.com/PolarSeeker/OpenSeeker
onyx.app/insights/best-open-source-llms-2026
fireworks.ai/blog/best-open-source-llms
github.com/MoonshotAI/Kimi-K2.5?tab=readme-ov-file
github.com/PolarSeeker/OpenSeeker
onyx.app/insights/best-open-source-llms-2026
fireworks.ai/blog/best-open-source-llms
github.com/MoonshotAI/Kimi-K2.5?tab=readme-ov-file
github.com/PolarSeeker/OpenSeeker
onyx.app/insights/best-open-source-llms-2026
fireworks.ai/blog/best-open-source-llms
github.com/MoonshotAI/Kimi-K2.5?tab=readme-ov-file
github.com/PolarSeeker/OpenSeeker
onyx.app/insights/best-open-source-llms-2026
fireworks.ai/blog/best-open-source-llms
github.com/MoonshotAI/Kimi-K2.5?tab=readme-ov-file
github.com/PolarSeeker/OpenSeeker
onyx.app/insights/best-open-source-llms-2026
The web,
organized for agentsOne API for search, crawling, and research agents
SearchAgent
latency
Fast450msAuto1sDeep\~10s
category
Full WebNewsCompaniesResearchPeopleFinancials
output
ResultsAnswerStructured
↓ 79% tokens
zai-org/GLM-5github.comGLM-5.1 is our next-generation flagship model for agentic engineering, with significantly stronger coding capabilities than its predecessor. It achieves state-of-the-art performance on SWE-Bench Pro and leads GLM-5 by a wide margin on NL2Repo (repo generation) and Terminal-Bench 2.0 (real-world terminal tasks).
MoonshotAI/Kimi-K2.5github.com15 trillion mixed visual and text tokens atop Kimi-K2-Base. It seamlessly integrates vision and language understanding with advanced agentic capabilities, instant and thinking modes, as well as conversational and agentic paradigms. ### Key Features - \\*Native Multimodality\\: Pre-trained on vision–language tokens,
Kimi K2.6 | OpenLM.aiopenlm.aiopen-source, native multimodal agentic model built through continual pretraining on approximately 15 trillion mixed visual and text tokens atop Kimi-K2-Base. It seamlessly integrates vision and language understanding with advanced agentic capabilities, instant and thinking modes, as well as conversational and agentic[
PolarSeeker/OpenSeekergithub.comwith open-source data \[...\] - \\2026.03.17\\ 🚀 We open-sourced OpenSeeker-v1 (all data and models). Using \\11.7K\\ training examples, we fine-tuned Qwen3-30B-A3B-Thinking-2507 and achieved scores of \\48.4%\\ on BrowseComp-ZH, \\29.5%\\ on BrowseComp, \\74.0%\\ on xbench-DeepSearch, and \\59.4%\\ on WideSearch.](https://github.com/PolarSeeker/OpenSeeker)
Google DeepMind Releases Gemma 4, Its ‘Most Capable’ Open-Source AI Models – SMBtechsmbtech.auGoogle DeepMind (the company’s flagship artificial intelligence research and development laboratory) has released Gemma 4, a new family of open-source AI models designed for advanced reasoning and agentic workflows, available under a commercially permissive Apache 2.0 licence.
Best Open Source LLMs in 2026: Rankings and Licensing Comparison | Onyx AIonyx.appThis guide covers the top 10 open-source and open-weight LLMs from the Onyx Open LLM Leaderboard, updated as of March 12, 2026.[
HiDream-ai/HiDream-O1-Imagegithub.comUnified Transformer (UiT) without external V \[...\] 48. \[...\] - 🚀 \\May 14, 2026:\\ We open-sourced \\\HiDream-O1-Image-Dev-2604\\*\ with its \prompt refiner\, tailored for text-to-image generation task.](https://github.com/HiDream-ai/HiDream-O1-Image/)[
Best Open Source LLMs in 2026: We Reviewed 7 Modelsfireworks.ai| Model | Release date | Params | Context window | Best for | On Fireworks | | --- | --- | --- | --- | --- | --- | \[...\] | DeepSeek v3.2 | December 1, 2025 | 671B total (37B active) | 128K - 163K tokens | Math reasoning, enterprise agents, and coding (96.0% GSM8K, 67.8% SWE-Bench) | Try in playground |](https://fireworks.ai/blog/best-open-source-llms)[
The Best Open-Source LLMs in 2026 - BentoMLbentoml.com\## DeepSeek-V4 \[...\] ## MiMo-V2.5-Pro# \[...\] ## GLM-5.1# \[...\] ## Gemma 4# \[...\] Reasoning: Kimi-K2.6, MiMo-V2.5-Pro \[...\] Coding assistants: DeepSeek-V4-Pro, Kimi-K2.6 \[...\] Agentic workflows: GLM-5.1, MiMo-V2.5-Pro \[...\] General chat: DeepSeek-V4-Pro \[...\] DeepSeek-V4-Pro \[...\] OpenClaw: Kimi-K2.](https://www.bentoml.com/blog/navigating-the-world-of-open-source-large-language-models)[
Best Open Source LLM Leaderboard 2026 - Onyx AIonyx.appBest Open Source \[...\] The definitive ranking of every major open source model — compared across reasoning, coding, math, software engineering, and instruction following benchmarks. \[...\] GLM-4.7 \[...\] Best Open Source Models \[...\] ## Open Source Model Benchmark Scores \[...\] K2.5 \[...\] ## Compare Open](https://onyx.app/open-llm-leaderboard)
RequestResponse


























Search API
Designed to make your Agent smarter

FINANCIAL DATA
COMPAW5
Powers their coding agents with smart token-efficient search across docs and repos
All the reports
your finance agent is looking forComprehensive coverage over all search verticals so your agent has the right context to solve its problem
Powers their coding agents with smart token-efficient search across docs and repos

Highlights: Off
90% TOKEN REDUCTION
#### Wikipedia - Boeing
airplanes , rotorcraft , rockets , satellites , and missiles worldwide. The company also provides leasing and product support services. Boeing is among the largest global aerospace manufacturers ; it is the fourth-largest defense contractor in the world based on 2022 revenue and is the largest exporter in the United States by dollar value. Boeing was founded in 1916 by William E. Boeing in Seattle , Washington.

Up to 25T tokens passed to models each week are concise excerpts powered by highlights
Token-efficient contents with highlights
Specialized model trained to extract the most relevant excerpts from the web
Try it out in our API Playground →

Up to 25T tokens passed to models each week are concise excerpts powered by highlights

"content": {
"companies": [
{
"company_name": "General Electric",
"ceo_name": "Larry Culp",
"founded_year": 1892
},
{
"company_name": "RTX",
"ceo_name": "Christopher T. Calio",
"founded_year": 2020
},
{
"company_name": "Boeing",
"ceo_name": "Kelly Ortberg",
"founded_year": 1916
}
]
}

Extracts enrichments as structured outputs across 70M+ companies with Exa
Low-latency to deep research
in one APIConfigure for fast results or deep research with structured outputs and grounded citations
Integrate Exa with your agent →

Extracts enrichments as structured outputs across 70M+ companies with Exa
Benchmarks
Highest quality search at every latency
Best-in-class accuracy for every search vertical and across the most challenging benchmarks
AccuracyLatencySearch Verticals
Exa Instant returns results in under 180ms, faster than any other search provider.
Accuracy
Exa leads across FRAMES, Tip-of-Tongue, and Seal0 — the most demanding retrieval benchmarks.
Latency
Exa Instant returns results in under 180ms, faster than any other search provider.
Search Verticals
Best-in-class across company search, people search, and code — not just general web queries.
Exa
Tavily
Brave
Parallel
Exa
Tavily
Brave
Parallel
Parallel
Brave
Tavily Ultra Fast
Exa Instant
3051ms
559ms
260ms
178ms
050010001500200025003000
Exa
Tavily
Brave
Parallel
Parallel
Brave
Tavily Ultra Fast
Exa Instant
3051ms
559ms
260ms
178ms
050010001500200025003000
Trusted by world-class teams

Exa powers all parts of Devin. We found that traditional search engine solutions were not enough. We chose Exa because it actually lets Devin search in a way that is natural for agents and returns content in a format that it can actually digest.

Walden Yan
Co-founder, Cognition

Exa is necessary for agents on HubSpot to supplement internal data with high quality, real-time people and company search. We found that other alternatives were slower, more expensive, and did not have comprehensive enrichment coverage.

Praty Sharma
AI Lead, HubSpot

We're entering a new era where software actually does the work. For that to be real, the agents need live intelligence behind them. Exa is where that comes from: current company and people data sourced from the web in real time.

Daniel Lereya
Chief Product and Technology Officer, monday.com
Exa consistently shows higher quality than other AI search providers we tested. What Exa is able to do in 1 search took our old provider multiple searches, taking 5 times as long.

David Loker
VP of AI, CodeRabbit

Standard signals like funding, job changes, and hiring are commoditized. Exa is how we find leads surfacing the triggers that actually predict a good prospect, then go deep on research before our agent engages.

Prabhav Jain
CEO at 11x

OpenRouter has a laser focus on bringing a great developer experience to all language models, and Exa is the best way we've found for grounding AI in the real world in a model-agnostic way.

Alex Atallah
CEO, OpenRouter

Exa's strong coverage and flexible API have been a key differentiator for us. Scientists trust our product further when the relevant papers they expect to see are available to them in the right workflows. The ability to customize search based on the use case easily, is important.

Naveed Janmohamed
CEO, Anara
When we saw companies show up in Exa that we hadn't found anywhere else, we knew it was our best option. It delivers, for a few dollars, what we'd been spending hundreds of thousands to assemble.

David Boskovic
CEO, Obvious

We use Exa for essentially all parts of our research—gathering sources, documenting, creating notes, and building briefs.

Christopher Varner
Head of Results Innovation

StackAI and Exa are the dream team…Exa is amazing at web search! We're proud to use Exa on our platform and to show our customers how web search can power AI agent workflows.

Bernard Aceituno
Co-founder and President, StackAI

Models are only as good as the data they're trained on, and Exa's search allowed us to get high quality data we couldn't find any other way.

Jonathan Frankle
Chief AI scientist, Databricks
Enterprise-grade security
and controls###### Zero Data Retention
Ensure true privacy and compliance with customized ZDR. All queries and data can be automatically purged based on your requirements.
###### SOC 2 Type II Certified
Our security framework maintains the highest level of compliance with industry standards. Safe information processing and access control.
###### Single-Sign On
A seamless, secure login experience for your entire team. Built-in team authentication and authorization management.
pura: loading…
☰
📷 🔍 Discover ●◼ –
Ready







