Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion docs/model-research/tuning.md
Original file line number Diff line number Diff line change
Expand Up @@ -245,7 +245,7 @@ engine for the model class, then tune.
runs; ExLlamaV3 and TensorFold are faster on short code and hold a smaller
window (section 7).
- Speed on one short request is not agent readiness. tqLLM decoded about 2.5x
faster than llama.cpp on the same GGUF and failed 8 of 11 multi-session checks
faster than llama.cpp on the same GGUF and passed at most 4 of 11 multi-session checks
(section 11); stock SGLang at 25 tok/s passed 11 of 11.
- The engine table is in [Engines](engines.md).

Expand Down
3 changes: 3 additions & 0 deletions oneshots/orrery/grades.tsv
Original file line number Diff line number Diff line change
Expand Up @@ -32,3 +32,6 @@ t5b2-qwen3.8-27b-vllm qwen3.8-27b-vllm 32000 length 32000 364.2 99.4 0 0 0 0 0 0
t5b3-qwen3.8-27b-vllm qwen3.8-27b-vllm 32000 length 32000 361.5 99.5 0 0 0 0 0 0 no SVG: length stop at 32,000 tokens, all thinking (five identical replies)
t5b4-qwen3.8-27b-vllm qwen3.8-27b-vllm 32000 length 32000 361.4 99.4 0 0 0 0 0 0 no SVG: length stop at 32,000 tokens, all thinking (five identical replies)
t5b5-qwen3.8-27b-vllm qwen3.8-27b-vllm 32000 length 32000 361.3 99.3 0 0 0 0 0 0 no SVG: length stop at 32,000 tokens, all thinking (five identical replies)
mx22-flashnext-heretic2-t5-oneshot-orrery-r1 lcpp6-flashnext-heretic2 140000 stop 33547 1513.4 22.3 0 0 0 0 0 0 Final model-supplied standalone SVG is malformed (mismatched tag, line 77). The harness also greedily concatenated two SVG blocks/prose: that packaging error was separated by extracting the final original codeblock verbatim, which still fails XML parsing. Initial valid background-only fragment is not substituted or graded.
t5nm1-lcpp6-ornith-122b-q4kxl lcpp6-ornith-122b-q4kxl 258048 stop 29646 1940.1 15.3 1 1 0 1 0 3 Parses and draws, but planets are left at the central origin instead of translated to arm ends. Only gears visibly move; outer markers stay fixed across sealed frames. Nominal period parameters are not functioning planetary orbits. Toothed wheels turn without credible meshing; no visible planetary day/night.
t5nm1-lcpp-flint-40b-q3kxl lcpp-flint-40b-q3kxl 258048 stop 30402 299.0 101.8 1 2 0 0 1 4 Renders with detached teeth and shadow halos; motion present; orbital periods 80/38/20/40/160/300s incorrectly order inner planets; no credible meshing pairs; offset black shadows rotate incorrectly. Independent original-source/frame review.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Original file line number Diff line number Diff line change
@@ -0,0 +1,45 @@
{
"case": "mx22-flashnext-heretic2-t5-oneshot-orrery-r1",
"finish_reason": "stop",
"first_token_s": 5.7,
"max_tokens": 140000,
"model": "lcpp6-flashnext-heretic2",
"reasoning_chars": 75955,
"reply_chars": 22823,
"started": "2026-10-23T04:28:15-0500",
"static": {
"svg_found": true,
"parses": false,
"parse_error": "not well-formed (invalid token): line 120, column 0",
"bytes": 21251,
"animateTransform": 0,
"animate": 16,
"animateMotion": 0,
"keyframes": 4,
"script": 1,
"prose_chars_outside_svg": 1596
},
"status": "ok",
"test": "5",
"timings": {
"cache_n": 0,
"prompt_n": 101,
"prompt_ms": 2666.171,
"prompt_per_token_ms": 26.397732673267324,
"prompt_per_second": 37.882041324431185,
"predicted_n": 33547,
"predicted_ms": 1507725.559,
"predicted_per_token_ms": 44.94501755798009,
"predicted_per_second": 22.249407260993447
},
"tok_s": 22.3,
"usage": {
"completion_tokens": 33547,
"prompt_tokens": 101,
"total_tokens": 33648,
"prompt_tokens_details": {
"cached_tokens": 0
}
},
"wall_s": 1513.4
}
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
44 changes: 44 additions & 0 deletions oneshots/orrery/t5nm1-lcpp-flint-40b-q3kxl/meta.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,44 @@
{
"case": "t5nm1-lcpp-flint-40b-q3kxl",
"finish_reason": "stop",
"first_token_s": 0.2,
"max_tokens": 258048,
"model": "lcpp-flint-40b-q3kxl",
"reasoning_chars": 97785,
"reply_chars": 14723,
"started": "2026-10-22T01:08:14-0500",
"static": {
"svg_found": true,
"parses": true,
"bytes": 14242,
"animateTransform": 13,
"animate": 2,
"animateMotion": 0,
"keyframes": 0,
"script": 0,
"prose_chars_outside_svg": 481
},
"status": "ok",
"test": "test5-orrery",
"timings": {
"cache_n": 0,
"prompt_n": 51,
"prompt_ms": 67.447,
"prompt_per_token_ms": 1.3224901960784314,
"prompt_per_second": 756.1492727623171,
"predicted_n": 30402,
"predicted_ms": 298794.899,
"predicted_per_token_ms": 9.828456267885924,
"predicted_per_second": 101.7453781900072
},
"tok_s": 101.8,
"usage": {
"completion_tokens": 30402,
"prompt_tokens": 51,
"total_tokens": 30453,
"prompt_tokens_details": {
"cached_tokens": 0
}
},
"wall_s": 299.0
}
207 changes: 207 additions & 0 deletions oneshots/orrery/t5nm1-lcpp-flint-40b-q3kxl/orrery.svg
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
44 changes: 44 additions & 0 deletions oneshots/orrery/t5nm1-lcpp6-ornith-122b-q4kxl/meta.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,44 @@
{
"case": "t5nm1-lcpp6-ornith-122b-q4kxl",
"finish_reason": "stop",
"first_token_s": 1.7,
"max_tokens": 258048,
"model": "lcpp6-ornith-122b-q4kxl",
"reasoning_chars": 87398,
"reply_chars": 15219,
"started": "2026-10-20T22:01:40-0500",
"static": {
"svg_found": true,
"parses": true,
"bytes": 14076,
"animateTransform": 20,
"animate": 0,
"animateMotion": 0,
"keyframes": 0,
"script": 0,
"prose_chars_outside_svg": 1146
},
"status": "ok",
"test": "test5-orrery",
"timings": {
"cache_n": 0,
"prompt_n": 59,
"prompt_ms": 1462.479,
"prompt_per_token_ms": 24.78777966101695,
"prompt_per_second": 40.34245961822358,
"predicted_n": 29646,
"predicted_ms": 1938393.583,
"predicted_per_token_ms": 65.3868639905549,
"predicted_per_second": 15.293591693653475
},
"tok_s": 15.3,
"usage": {
"completion_tokens": 29646,
"prompt_tokens": 59,
"total_tokens": 29705,
"prompt_tokens_details": {
"cached_tokens": 0
}
},
"wall_s": 1940.1
}
Loading
Loading