LuminaXspace weekly report
Ten posts selected from the week’s strongest AI news, credible reports, leaks and model progress—shown with their original X media and engagement snapshot.
• Expected to launch in August • Potentially more than 1 trillion total parameters • Will likely remain Open Weight • Should retain its 1M token Context Window • Built for long running autonomous coding and agent tasks This model should blow Kimi K3 and Qwen 3.8 out of the water. Do you think it will close the gap finally and challenge Anthropic/OpenAI?
Anthropic’s next flagship model is rumoured to be hiding behind the name "Honeycomb": • Briefly appeared inside Cursor before being removed • Reportedly supports a 1M token context window • Included an “xhigh” reasoning effort mode • May be targeting release before the end of July • Expected to focus heavily on long-running agents and advanced coding • Should be closer to Anthropic’s Fable/Mythos tier Honeycomb has not been officially confirmed as Claude Opus 5, so the final model name and release timing could still change. Will Opus 5 be a model you switch to?
Google’s next flagship model is said to have missed internal performance targets again: • Coding performance reportedly remains a key weakness • The release is now months behind its original schedule • Google is still working to close the gap with Claude, GPT and Grok • No new launch date has been confirmed Alphabet shares fell more than 4% as concerns grew over Google’s position in the AI race. Do you think Gemini 3.5 Pro will be worth the wait or just keep getting pushed back?
The latest claim points to a possible July 20/21 launch: • Should still has a 1M-token context window • Capability is rumoured to sit much closer to Fable 5 Opus 5 is supposedly the “Honeycomb EAP” sighting inside Cursor, which briefly showed 1M context, xhigh reasoning and an Opus 4.8 fallback which im guessing is because it is a major step up in ability. Is Opus 5 a model your looking out for or do you want Fable 5.1?
The current model watchlist (I can't believe my eyes): • Claude Opus 5 (possibly in July but not 100% yet) • GPT-6 (August/September rumour, still unverified) • Claude Fable 5.1 (expected, but not 100% yet) • Grok 2T (initial training finishes next week) • GLM-5.3 • Gemini 3.5 Pro (delayed again, if it doesn't launch in august google are done) • Gemini 3.6 Flash (already being a/b tested) • DeepSeek V4 GA • Qwen 3.8 • MiniMax M3.1 • Composer 3 I can only imagine that if half of these land, the frontier leaderboard will change every week (maybe every few days). Which model are you most excited for out of the list, mine would be GPT-6?
Multiple people are seeing it live on grok right now: • Expert mode: grok-v9-4p5-chat-expert • Heavy mode: grok-v9-4p5-multi (up to 4 agents) • Already available for SuperGrok and X Premium+ users • Broader access expected in the next few days to 1–2 weeks It's weird seeing xAI shipping 4.5 without a big announcement. Have you got access yet?
Kimi K3 has reached the top of the Frontend Code Arena. • 2.8 trillion total parameters • 1M token context window • And it beats Claude Fable 5😂 The main issue with it is the speed of it, it takes way too long to do what you want. What are your opinions on Kimi K3 so far?
Deepseek is reportedly seeking another enormous funding round: • New discussions could value DeepSeek at around $71 billion • Funding would support a dedicated DeepSeek data centre • The company intends to acquire significantly more AI chips • Future development is reportedly focused on autonomous AI agents • Agent systems require far more inference compute than ordinary chat models DeepSeek has not announced a V5 model, but the infrastructure expansion suggests its next generation could be far more agent-focused. Are you looking forward to Deepseek V5?
Some users are now seeing Kimi K3 inside the beta model selector: • Kimi K3 reportedly visible on Moonshot’s beta website • Access appears limited to selected accounts • The model is not yet available on the main Kimi platform • Rumours suggest a 2–3T parameter architecture • Official benchmarks, pricing and context limits remain unknown This follows the leaked July 15 (Chinese Time) Kimi K3 launch promotion, suggesting the full release could begin within hours. Do you think Kimi K3 be the strongest Chinese frontier AI model?
