Spaces:
Running
Running
Commit History
28 days ago
-
Restore pure scientific benchmark suite (remove synthetic interactive playground)
Verified signature262df17 -
Add Section 05: Interactive Whole-Repository Synthesis Playground for all 4 Ael models
Verified signature1035e64
29 days ago
-
Update 40B flagship model name to Ael-Pro-40B across Benchmark Space
Verified signaturef85eb26 -
Apple + ChatGPT UI layout: Bento grid, segmented controls, capsule bars, 4 Ael models
Verified signature5041987 -
Apple + ChatGPT UI theme: Pure Black (#000000) dark mode & Light Gray (#F5F5F7) bright mode
Verified signature32d10a4 -
Update data/ael_4models_multi_million_benchmarks.json: Clean OpenAI-grade 4-model Ael benchmark report (2.15M-2.77M tokens)
Verified signaturea6d8dd0 -
Update style.css: Clean OpenAI-grade 4-model Ael benchmark report (2.15M-2.77M tokens)
Verified signature89566a6 -
Update index.html: Clean OpenAI-grade 4-model Ael benchmark report (2.15M-2.77M tokens)
Verified signature70e6bbe -
Update README.md: Clean OpenAI-grade 4-model Ael benchmark report (2.15M-2.77M tokens)
Verified signatureb0bbbd4
-
Upload audited_results/isom_r2_512k_retrieval_results.json with huggingface_hub
Verified signature0c4eafb -
[Telemetry] Add audited ISOM-R1-Coder-1.5B V2 spectral recall benchmark results
Verified signature095afe1 -
Add ISOM-R1 V2 Multi-Scale Spectral Verification download card & telemetry
Verified signaturef1ae2f8 -
Upload data/isom_r1_v2_mathematical_verification.json with huggingface_hub
Verified signature8e7ac3f
-
Add ISOM-R1 Mathematical Verification and ISOM-R2 512K Retrieval download cards & telemetry
Verified signature502e0ff -
Add ISOM-R2 512K Retrieval Benchmark JSON results (Tesla T4, 97.59% KV reduction)
Verified signature8a2ce10 -
Add ISOM-R1 Mathematical & Spectral Verification JSON telemetry (100K steps, SO(2) isometry)
Verified signatureb5fc8b4
-
Update Space index.html with Kaggle Tesla T4 verified benchmark card and badges
Verified signaturebc227a0 -
Add verified Kaggle Tesla T4 hardware benchmark JSON telemetry
Verified signaturec058393 -
docs: scrub benchmark reduction claims for 16B and 40B models
Verified signaturedfca9d2 -
docs: update to ISOM-R1 sovereign signature with non-endorsement lineage notes
Verified signature880f68d -
ui: align model cards and lineup names
Verified signature4f79d0a -
docs: update to ISOM-R1 sovereign signature with non-endorsement lineage notes
Verified signaturee056b69 -
ui: update all instruct model titles and card names to -Instruct
Verified signature22f8479 -
docs: update to ISOM-R1 sovereign signature with non-endorsement lineage notes
Verified signature8ef2ff0 -
ui: update links to new sovereign ISOM-R1 repository names
Verified signaturef6aae92 -
docs: update to ISOM-R1 sovereign signature with non-endorsement lineage notes
Verified signaturec06ca01 -
ui: align ISOM-R1 lineup table and clean lineage documentation
Verified signature8e27452 -
docs: update to ISOM-R1 sovereign signature with non-endorsement lineage notes
Verified signatureb5adbff -
ui: update ISOM-R1 sovereign signatures and non-endorsement notices
Verified signature3725e60 -
docs: update to ISOM-R1 sovereign signature with non-endorsement lineage notes
Verified signaturec937e11 -
ui: add 5-model unified hybrid lineup, base lineage, clean contacts and benchmarks
Verified signature6b63519 -
docs: add unified hybrid signature lineup and base model lineage notes
Verified signaturee7bd5e4 -
docs: rewrite README removing unsubstantiated HumanEval/ProofWriter claims
Verified signature2b1becc -
fix: replace all HumanEval/ProofWriter claims with real audited 128K Qwen2.5-Coder data
Verified signature72a484a -
fix: remove all unsubstantiated +20.7%/+30.0% HumanEval/ProofWriter claims; clean Space
Verified signature4d4b84a -
fix: remove +20.7% HumanEval & +30.0% Logic claims, remove HumanEval/ProofWriter tabs
Verified signature5af1001 -
fix: remove unsubstantiated +20.7% HumanEval & +30.0% Logic claims; remove HumanEval/ProofWriter tabs
Verified signature5b02052 -
docs: crystal-clear formatting, ASCII math blocks, and unified ecosystem
Verified signature1dc95c9 -
docs: complete multi-model ISOM portfolio, 128K Qwen2.5-Coder, and mathematical formulations
Verified signaturef1ff4e8 -
feat: add ISOM-Qwen2.5-Coder 128K benchmark card, profiler logic, and download links
Verified signature0e1dd7b -
data: add audited 4-pillar hardware benchmark for ISOM-Qwen2.5-Coder-1.5B-Instruct
Verified signature2b5c706
-
docs: add LinkedIn research integration contact link
Verified signaturee78c8a1
-
Document ISOM IQ Accelerator and unified model portfolio
Verified signature068667f -
Feature ISOM IQ Accelerator and ISOM-130M model integration
Verified signaturedc57650 -
Update space: pure ISOM branding (zero Qwen, zero Samba)
Verified signature0b38964 -
Deploy 100% ISOM benchmark suite with verified ISOM data checkpoints
Verified signature6c66de7 -
Remove redundant metadata file
Verified signaturee29d48f -
Rebrand to ISOM (Isometric State Operator Manifold): Isom-Qwen & Isom-Falcon
Verified signature4b0684d -
Update Test #6 8K 75% depth to Near Match (99%) with audited BPE tokenizer splitting note
Verified signature2bfc7a9 -
Apply CC BY-NC-ND 4.0 license across benchmark space
Verified signature30500de