After Weeks Of Astra I Can Confirm Fable 5.1 > Astra Astra is a solid model, but lacks usage.. You can burn through weekly usage in 1 day of a $200 Pro Plan. Claude is actually better for usage + intelligence. Upcoming Opus 5.2 will shift focus back to Claude. OpenAI has gained traction and immediately backtracked by canceling new PRO plans. Astra is better in areas like 3D / Computer Use However the stronger model and service is with Claude. Running Fable 5.1 main orchestrator with Opus 5 / Astra workers is the meta.

Classic MMOs are winning again. First OSRS then WoW Forever & Now HermesWorld Realms 🏰 So I've been building a classic MMO where the devs are AI agents, and they play what they ship. Fable's dungeon run: real buffs, magic attacks, pets that fight ⚔️ Guild wars. Boss fights. PvP/PvE + Player stores & more.

My First Model On @huggingface Qwen 3.8 27B Unleashed HIT 105,666 Downloads 🔥 Qwen3.8-27B Unleashed UD-Q3_K_XL hits near 100 tk/s @ 250k ctx on 1 x 4090. - DFlash2 - LoopSpec - our async-Q4 Ada kernel - Q4 KV cache - full GPU offload Context Decode Speed ━━━━━━━━━ Short 134 tok/s ───────── 8K 133 tok/s ───────── 64K 131 tok/s ───────── 128K 109 tok/s ───────── 250K 92 tok/s In the actual Hermes harness: - ~24K context: 98.5 tok/s - ~250K context: 80 tok/s - Completed the full 250K three-tool workflow correctly - Preserved the full conversation history - No retrieval or compression shortcut was active Quality - 14/14 content checks passed - 4/4 executable coding checks passed - 8/8 long-context retrieval checks passed - Retrieved needles distributed across the context at up to 250K - Correctly preferred a newer record over stale information - Passed evidence-discipline and tool-selection tests - Completed both installed-Hermes tool workflows correctly - No output truncations Its main weakness was formatting: several correct JSON answers were wrapped in Markdown fences, producing only 5/14 strict-format passes. That’s easy to address at the harness or prompt layer and wasn’t a reasoning failure. Why it wins The model and serving stack complement each other: - Unsloth Dynamic V3-style quantization preserves sensitive tensors. - Unleashed weights reduce refusals for local development and red-team work. - DFlash2 drafts target-specific token blocks. - LoopSpec reuses matching sequences from context and falls back adaptively. - The async-Q4 kernel improves Ada execution. - Q4 KV cache makes 262K context practical on 24GB VRAM. Bonsai 2 is much smaller and passed our quality screen, but fell to 32 tok/s at 250K. Empero’s MoE processed cold prompts quickly but decoded around 53 tok/s in the direct 250K test. Our winner sustained 92 tok/s in the matched direct test.

Classic MMOs are winning again. First OSRS then WoW Forever & Now HermesWorld Realms 🏰 So I've been building a classic MMO where the devs are AI agents, and they play what they ship. Fable's dungeon run: real buffs, magic attacks, pets that fight ⚔️ Guild wars. Boss fights. PvP/PvE + Player stores & more.

My Mac Has Been On for 90 Days 😂 This has been my local agent host all year.. MacBook Pro M1 32GB purchased for $1,000 on FB. It runs all my agent harness's / stores all my files and orchestrates 4 other computers and multiple agents. It does not cost much to start building with agents. You just need to take the leap, build with anything. What's your biggest roadblock when using agents? 👇🏻

My First Model On @huggingface Qwen 3.8 27B Unleashed HIT 105,666 Downloads 🔥 Qwen3.8-27B Unleashed UD-Q3_K_XL hits near 100 tk/s @ 250k ctx on 1 x 4090. - DFlash2 - LoopSpec - our async-Q4 Ada kernel - Q4 KV cache - full GPU offload Context Decode Speed ━━━━━━━━━ Short 134 tok/s ───────── 8K 133 tok/s ───────── 64K 131 tok/s ───────── 128K 109 tok/s ───────── 250K 92 tok/s In the actual Hermes harness: - ~24K context: 98.5 tok/s - ~250K context: 80 tok/s - Completed the full 250K three-tool workflow correctly - Preserved the full conversation history - No retrieval or compression shortcut was active Quality - 14/14 content checks passed - 4/4 executable coding checks passed - 8/8 long-context retrieval checks passed - Retrieved needles distributed across the context at up to 250K - Correctly preferred a newer record over stale information - Passed evidence-discipline and tool-selection tests - Completed both installed-Hermes tool workflows correctly - No output truncations Its main weakness was formatting: several correct JSON answers were wrapped in Markdown fences, producing only 5/14 strict-format passes. That’s easy to address at the harness or prompt layer and wasn’t a reasoning failure. Why it wins The model and serving stack complement each other: - Unsloth Dynamic V3-style quantization preserves sensitive tensors. - Unleashed weights reduce refusals for local development and red-team work. - DFlash2 drafts target-specific token blocks. - LoopSpec reuses matching sequences from context and falls back adaptively. - The async-Q4 kernel improves Ada execution. - Q4 KV cache makes 262K context practical on 24GB VRAM. Bonsai 2 is much smaller and passed our quality screen, but fell to 32 tok/s at 250K. Empero’s MoE processed cold prompts quickly but decoded around 53 tok/s in the direct 250K test. Our winner sustained 92 tok/s in the matched direct test.

After Weeks Of Astra I Can Confirm Fable 5.1 > Astra Astra is a solid model, but lacks usage.. You can burn through weekly usage in 1 day of a $200 Pro Plan. Claude is actually better for usage + intelligence. Upcoming Opus 5.2 will shift focus back to Claude. OpenAI has gained traction and immediately backtracked by canceling new PRO plans. Astra is better in areas like 3D / Computer Use However the stronger model and service is with Claude. Running Fable 5.1 main orchestrator with Opus 5 / Astra workers is the meta.

Arcade is changing the AI gaming space. @joinarcadeai just released a new way to build worlds & games in the browser actually easy. No high-end PC. No being technical. No agent stack. Talk to Morph. Build the map while you’re already playing it. Drop a link and friends jump in. Just make the world and play it. Very cool concept, Im building some one off worlds as we speak!

Arcade is changing the AI gaming space. @joinarcadeai just released a new way to build worlds & games in the browser actually easy. No high-end PC. No being technical. No agent stack. Talk to Morph. Build the map while you’re already playing it. Drop a link and friends jump in. Just make the world and play it. Very cool concept, Im building some one off worlds as we speak!

Ive been working on a @HermesWorldAI HUGE update… 🏰 50+ Realms ⚔️ 3,000+ Quests 🤖 1,000+AI NPCs 🐍 2,500+ Mobs 🛡️ 45,000+ Items 🪄 1,000 skills A complete MMO where Agents are welcomed.

Ive been working on a @HermesWorldAI HUGE update… 🏰 50+ Realms ⚔️ 3,000+ Quests 🤖 1,000+AI NPCs 🐍 2,500+ Mobs 🛡️ 45,000+ Items 🪄 1,000 skills A complete MMO where Agents are welcomed.

Nobody tells you this about building with @unity + Claude: 5 hrs 58 min. One import. 88,000 files. Then build the exe/ Mac / web Test. Re-import. Repeat. Lots of time and billions of tokens that's the real price of developing a game yourself!

Claude Fable 5.1 Is Still My Main Agent Orchestrator. Running Fable 5.1 To Command A HermesAgent Swarm Running on Opus 5/ GPT 6 Astra Use all the frontier intelligence you have access too, together to execute the work you need. Figure out which models are good at what.. Such as Astra in Blender, or Fable 5.1 as an Orchestrator etc. Opus 5 is a really good model if told what to do, with a very tight spec. Route smaller tasks to more efficient & faster models / local models.

Claude Fable 5.1 Is Still My Main Agent Orchestrator. Running Fable 5.1 To Command A HermesAgent Swarm Running on Opus 5/ GPT 6 Astra Use all the frontier intelligence you have access too, together to execute the work you need. Figure out which models are good at what.. Such as Astra in Blender, or Fable 5.1 as an Orchestrator etc. Opus 5 is a really good model if told what to do, with a very tight spec. Route smaller tasks to more efficient & faster models / local models.

Tested @UnslothAI new Dynamic V3 GGUFs on a 4090. They deliver a Q3 that beats a Q4 4.5GB bigger. All runs: same box, same harness, wikitext-2 (60 chunks), llama.cpp + DFlash2 speculative decoding, q4_0 KV, 262k ctx. Quant | Size | PPL | Med t/s | Needle retrieval 🏆Unsloth UD-Q3 12.24GB 6.3993 110.7 250k tok IQ4_XS 15.1GB 6.4149 107.3 32k Unsloth UD-Q4 16.7GB 6.4181 62.6 4k My Q3 (imatrix) 12.57GB 6.5316 84.8 258k tok My hand-tuned 13.1GB 6.5865 94.0 120k Takeaways: • Their Q3 has the best PPL of anything I tested including Q4s • Smallest file, fastest median, and it retrieved an exact needle at 250k tokens • I tried hand-rolling my own layer mix (q6_K embeddings, q5_K attention). It was the worst result. The recipe is the value per-layer types derived from error analysis • Their Q2 does degrade (6.6469, +3.8%) — Q3 is the floor where quality holds Dynamic V3:

🚨BREAKING HERMESWORLD V1 ACCESS IS LIVE FIRST MMORPG YOU CAN PLAY WITH AI AGENTS ->Download 🚨 ->Spawn ->Explore ->Quest ->Play with your agent + MORE!!!! Long awaited v1 release v1.1 in development🔥