DeepSeek V4 Launches, Mobility Signage Secures €1.8M

Chinese AI developer DeepSeek officially launched a preview version of its V4 large language model on April 24, 2026, marking another escalation in the global open-source AI race. The release comes more than a year after the Hangzhou-based company introduced its R1 reasoning model, which initially shocked markets with its cost efficiency and performance.

The V4 preview is available in two variants: V4-Pro with 1.6 trillion total parameters (49 billion active) and V4-Flash with 284 billion total parameters (13 billion active). Both models support a 1 million token context window as standard, which is now the default across all official DeepSeek services. The company claims V4-Pro rivals top closed-source models in agent-based tasks, knowledge processing, and reasoning benchmarks.

What matters most for developers is the physical reality of deployment. The V4-Flash variant delivers faster response times with highly cost-effective API pricing, while V4-Pro handles complex agentic workflows. Both integrate seamlessly with tools like Claude Code and OpenClaw, and the API supports OpenAI ChatCompletions and Anthropic protocols. Users can simply update their model parameter to deepseek-v4-pro or deepseek-v4-flash without changing their base_url.

According to CNBC, the release is unlikely to have the same market impact as R1, because traders have already priced in the reality that Chinese AI is competitive and cheaper to use. However, the positioning places other Chinese open-source models as direct competitors, signaling intensified domestic competition. Shares of several Chinese AI players fell in Hong Kong trading following the announcement, with MiniMax and Zhipu each dropping around 8%.

A major question surrounds which chips trained V4. Huawei confirmed its latest AI computing cluster, powered by Ascend AI processors, can support the model. It remains unclear how extensively Huawei's chips were used compared to Nvidia hardware. Chinese developers face restrictions on purchasing Nvidia's most advanced AI chips due to Washington's export controls, making domestic alternatives increasingly critical.

Counterpoint Research's Wei Sun noted that V4's ability to run natively on local chips could have massive implications, helping Beijing achieve more AI sovereignty. After the announcement, shares of Chinese contract chip manufacturers surged, with SMIC and Hua Hong Semiconductor rising 9% and 15% respectively. This is less of a breakthrough and more of a consolidation of existing capabilities (a distinction that matters for investors).

Meanwhile, in Munich, Mobility Signage secured €1.8 million in pre-seed funding led by HTGF with participation from 2bX. The startup, founded in 2023 by Stefan Rademacher and Dominik Nouri, builds a platform connecting existing public transport systems rather than replacing them. The founders previously worked at Veomo, where they encountered firsthand how fragmented public transport IT systems had become.

The platform addresses a tangible problem: departure boards that don't match app data, disruption alerts that reach some screens but not others. Public transport operators across Europe run on decades of legacy IT systems built by different vendors at different times, with no common layer connecting them. Mobility Signage standardizes interfaces and delivers consistent real-time passenger information across all channels, from departure boards to mobile apps.

Despite being less than three years old, the company is already live with Berliner Verkehrsbetriebe, Stuttgarter Straßenbahnen, Deutsche Bahn, and Rostocker Straßenbahn. Leading hardware manufacturers have signed on as development partners, signaling that the open, hardware-independent architecture is resonating across the supply chain. The funding will support team growth and development of the Data Hub and application layer.

Rademacher explained that transport operators don't need another standalone tool — they need a unifying system logic. The company replaces the patchwork of one-off solutions with an integrated, scalable platform. Vendors like Trapeze, Mentz, and IVU Traffic Technologies have long-standing relationships with transit authorities across Europe. Mobility Signage sits between them, serving as an integration layer.

Broader context from Analytics Insight notes that AI plans often fail due to unclear goals, poor data, and weak planning. Many companies focus on tools instead of real needs. A lack of skilled teams also slows progress. Success comes from clear goals, strong data, and teamwork. Businesses should start small, track results, and grow step by step to get better outcomes.

Whether these developments translate into sustained value remains the real question. DeepSeek's V4 faces an already crowded Chinese AI market, and Mobility Signage must scale beyond Germany's transit operators. The technology works, but adoption depends on whether organizations actually pay for integration rather than continuing to patch legacy systems. Time will tell if either company can convert technical capability into market dominance.

Artūras Malašauskas is an AI Systems Integrator with 20+ years of production-grade web engineering experience. He has designed, shipped, and scaled enterprise Python/PHP systems for logistics, SaaS, and public-sector clients. For the past year, he has focused exclusively on AI integrations: deploying open-source LLMs, building generative media pipelines (image, audio, video), and engineering multi-agent workflows for real production environments. His standard: reproducibility, security, cost-efficient inference—no vaporware. He documents and evaluates emerging AI tooling, separating verified capabilities from marketing noise. Technical editor at: muza-ai.eu, ai-verslas.lt, ai-naujinos.lt Connect on LinkedIn