basaltlabsai/monolith-1.0
Updated • 33
BuiDoan
BuiDoan
AI & ML interests
None yet
Recent Activity
liked a model about 13 hours ago
basaltlabsai/monolith-1.0 upvoted an article 1 day ago
Welcome Inkling by Thinking Machines liked a model 2 days ago
thinkingmachines/InklingOrganizations
liked a
model about 13 hours ago
upvoted an article 1 day ago
Article
Welcome Inkling by Thinking Machines
- +2
burtenshaw, merve, pcuenq, ariG23498
• • 90 liked a
model 2 days ago
liked 3
models 8 days ago
IDEA: Bitnet 1.58 (a4.8) version in future variants would be so incredible!
❤️ 4
3
#10 opened 12 days ago
by 
apiarium
commented on Qwen3.6-27B-FP8 on One RTX 6000 Ada: Fast TTFT, 314 tok/s Decode Generation [Benchmark] 12 days ago
- Quantization: Utilize 4-bit quantization (e.g., AWQ, GPTQ) to optimize GPU memory usage, significantly improving TTFT (Time To First Token) and increasing total throughput (token generation rate).
- Inference Acceleration: Leverage MTP to enhance inference efficiency, achieving a performance boost of 1.3x–2.0x.
- Context Management: Configure the --max-model-len parameter (typically 32k–64k) to expand model capability; note that this significantly impacts KV Cache memory footprint, requiring careful VRAM allocation management.
reacted to ginigen-ai's post with ❤️ 18 days ago
Post
10445
🧠 Does your LLM know when it's about to be wrong?
Most leaderboards measure accuracy. We measure metacognition — whether a model catches its own errors. Benchmark + leaderboard + adapters, all open. 🎉
The surprise: even a K-AI #1 model (JGOS-31B-Citizen) is the strongest on multiple-choice traps (trap_rate 0.005 — ~2 misses in 400) yet blind to its own free-form mistakes (self-confidence AUROC = 0.5, pure random). A tiny base-frozen adapter recovers that signal.
Two independent axes (never compared across a row): ① trap_rate — does it fall for tempting trap options? (lower = stronger) ② adapter gain Δ — how much a lightweight adapter catches errors the model itself misses. (higher = more adapter value)
What's open: 📊 300+100 trap problems (each with a hidden trap + TICOS type) 🏆 24-model leaderboard 🧩 11 per-model adapters — adapters, NOT fine-tunes (base stays frozen; the adapter just reads the hidden state → P(wrong))
Submit any HF model → auto-scored daily at 09:00 KST and added to the board.
🏆 Leaderboard → ginigen-ai/Metacognition-Leaderboard-Space
📊 Benchmark → ginigen-ai/Metacognition-Bench
🧩 Adapters → FINAL-Bench/metacognition-adapters-6a42c032e6beb803dd032961
📊 Article → https://huggingface.co/blog/ginigen-ai/metacognition
Benchmark by ginigen-ai · Adapters by FINAL-Bench (Darwin/Chimera platform + AETHER metacognition tech).
Most leaderboards measure accuracy. We measure metacognition — whether a model catches its own errors. Benchmark + leaderboard + adapters, all open. 🎉
The surprise: even a K-AI #1 model (JGOS-31B-Citizen) is the strongest on multiple-choice traps (trap_rate 0.005 — ~2 misses in 400) yet blind to its own free-form mistakes (self-confidence AUROC = 0.5, pure random). A tiny base-frozen adapter recovers that signal.
Two independent axes (never compared across a row): ① trap_rate — does it fall for tempting trap options? (lower = stronger) ② adapter gain Δ — how much a lightweight adapter catches errors the model itself misses. (higher = more adapter value)
What's open: 📊 300+100 trap problems (each with a hidden trap + TICOS type) 🏆 24-model leaderboard 🧩 11 per-model adapters — adapters, NOT fine-tunes (base stays frozen; the adapter just reads the hidden state → P(wrong))
Submit any HF model → auto-scored daily at 09:00 KST and added to the board.
🏆 Leaderboard → ginigen-ai/Metacognition-Leaderboard-Space
📊 Benchmark → ginigen-ai/Metacognition-Bench
🧩 Adapters → FINAL-Bench/metacognition-adapters-6a42c032e6beb803dd032961
📊 Article → https://huggingface.co/blog/ginigen-ai/metacognition
Benchmark by ginigen-ai · Adapters by FINAL-Bench (Darwin/Chimera platform + AETHER metacognition tech).
- 11 replies
liked a
model 21 days ago
liked a
model 22 days ago
liked 3
models 23 days ago
reacted to sequelbox's post with 🔥 25 days ago
Post
7444
NEW RELEASE: Esper 4 is here for Qwen 3.6 27b, along with our new datasets!
- NEW DATASET: Titanium 4 maximizes DevOps and architecture helpfulness, powered by high-difficulty agentic-focused DevOps and architecture data generated with DeepSeek-V4-Pro!
- NEW DATASET: Mitakihara 2 brings AI coding and expertise data for AI development, research, deployment, interpretability, operation and experimentation!
- Improved coding performance: challenging agentic coding queries from Tachibana 4 allow Esper 4 to tackle harder coding tasks across a variety of languages!
GET ESPER 4: ValiantLabs/Qwen3.6-27B-Esper4
Get the datasets for your own training:
sequelbox/Titanium4-DeepSeek-V4-Pro
sequelbox/Mitakihara2-DeepSeek-V4-Pro
sequelbox/Tachibana4-DeepSeek-V4-Pro
We've been working hard on Esper 4 - it's so exciting to finally bring it to everyone! We hope it helps you build.
We'll be expanding Esper 4 to more models as funding allows - donate for more, faster, better models and datasets: sequelbox/SupportOpenSource
The revolution is coming - we're here to fight for AI you can use and build on your own computer, not a giant corporation charging you for access at their discretion. We've seen what OpenAI, Anthropic, and the ultra-rich taking charge of the AI future looks like, and it's already very clear you won't like living in it. Choose a different future while you still can.
Open source must win.
More to come soon!
love, always,
allegra
- NEW DATASET: Titanium 4 maximizes DevOps and architecture helpfulness, powered by high-difficulty agentic-focused DevOps and architecture data generated with DeepSeek-V4-Pro!
- NEW DATASET: Mitakihara 2 brings AI coding and expertise data for AI development, research, deployment, interpretability, operation and experimentation!
- Improved coding performance: challenging agentic coding queries from Tachibana 4 allow Esper 4 to tackle harder coding tasks across a variety of languages!
GET ESPER 4: ValiantLabs/Qwen3.6-27B-Esper4
Get the datasets for your own training:
sequelbox/Titanium4-DeepSeek-V4-Pro
sequelbox/Mitakihara2-DeepSeek-V4-Pro
sequelbox/Tachibana4-DeepSeek-V4-Pro
We've been working hard on Esper 4 - it's so exciting to finally bring it to everyone! We hope it helps you build.
We'll be expanding Esper 4 to more models as funding allows - donate for more, faster, better models and datasets: sequelbox/SupportOpenSource
The revolution is coming - we're here to fight for AI you can use and build on your own computer, not a giant corporation charging you for access at their discretion. We've seen what OpenAI, Anthropic, and the ultra-rich taking charge of the AI future looks like, and it's already very clear you won't like living in it. Choose a different future while you still can.
Open source must win.
More to come soon!
love, always,
allegra
- 4 replies
liked 5
models 27 days ago
poolside/Laguna-M.1
Text Generation • 226B • Updated • 3.78k • 120
nex-agi/Nex-N2-Pro
Text Generation • 397B • Updated • 2.13k • 368
jedisct1/Nex-N2-mini-mlx-OptiQ-4bit-MTP
Text Generation • 6B • Updated • 648 • 6
cyankiwi/Nex-N2-mini-AWQ-INT4
Text Generation • 37B • Updated • 14.8k • 10
nex-agi/Nex-N2-mini
Text Generation • 35B • Updated • 6.76k • 284