Benchmark Almanac

FPL 2026 Agentic FPGA Backend Optimization Competition

AER 4.29 tops its leaderboard
Not in model cards

Teams improve the maximum clock frequency of routed FPGA designs while preserving behavior. The organizers rank teams across seven designs after accounting for runtime and model cost.

The organizers scored 24 teams on seven final designs. MEEMAR reported language model spending on four designs. Teams could also use deterministic methods.

Released
February 2026
Kind
Electronics
Level
Expert
Scored
Checked automatically
Answers
Agent task
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Code → Code, Actions
Interaction
Agentic
Environment
Terminal
Refresh
One fixed set
Run by
Anyone
In views
Agents
24
Entries on its own leaderboard, 0 models we track

AER tops the leaderboard with 4.29

Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.

  1. 1 AER 4.29
  2. 2 RouteAgents 5.00
  3. 3 ESCA 6.29
  4. 4 Fu_Do_Fdo 7.43
  5. 5 MEEMAR 7.71
  6. 6 Fmaximizers 8.86
  7. 7 Novi_CAD 9.57
  8. 8 CUFO 10.1
  9. 9 BSC-ANO 10.9
  10. 10 Elios_Lab 11.3
  11. 11 UTDA_FPGA 11.4
  12. 12 Synthesis_Sages 12.1
  13. 13 critpath 12.6
  14. 14 LogicFlow 12.6
  15. 15 Net-Surgeon 13.1
  16. 16 ParaGato_Labs 13.6
  17. 17 Morpheus 15.6
  18. 18 FLY_FPGA 16.0
  19. 19 nagar 16.3
  20. 20 DaKaiMen 16.6
  21. 21 OOM_Slayers 16.7
  22. 22 NanoLLM 18.0
  23. 23 Berkeley_Gobblers 18.6
  24. 24 Microprocessor_Lab 22.4

Source: Official FPL 2026 Agentic FPGA Backend Optimization Competition leaderboard

#EntryScoreSettingSubmitted byDateLinkSource
1 AER 4.29 AER github.com Official leaderboard
2 RouteAgents 5.00 RouteAgents github.com Official leaderboard
3 ESCA 6.29 ESCA github.com Official leaderboard
4 Fu_Do_Fdo 7.43 Fu_Do_Fdo github.com Official leaderboard
5 MEEMAR 7.71 MEEMAR github.com Official leaderboard
6 Fmaximizers 8.86 Fmaximizers Official leaderboard
7 Novi_CAD 9.57 Novi_CAD Official leaderboard
8 CUFO 10.1 CUFO Official leaderboard
9 BSC-ANO 10.9 BSC-ANO Official leaderboard
10 Elios_Lab 11.3 Elios_Lab Official leaderboard
11 UTDA_FPGA 11.4 UTDA_FPGA Official leaderboard
12 Synthesis_Sages 12.1 Synthesis_Sages Official leaderboard
13 critpath 12.6 critpath Official leaderboard
14 LogicFlow 12.6 LogicFlow Official leaderboard
15 Net-Surgeon 13.1 Net-Surgeon Official leaderboard
16 ParaGato_Labs 13.6 ParaGato_Labs Official leaderboard
17 Morpheus 15.6 Morpheus Official leaderboard
18 FLY_FPGA 16.0 FLY_FPGA Official leaderboard
19 nagar 16.3 nagar Official leaderboard
20 DaKaiMen 16.6 DaKaiMen Official leaderboard
21 OOM_Slayers 16.7 OOM_Slayers Official leaderboard
22 NanoLLM 18.0 NanoLLM Official leaderboard
23 Berkeley_Gobblers 18.6 Berkeley_Gobblers Official leaderboard
24 Microprocessor_Lab 22.4 Microprocessor_Lab Official leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Engineering and CAD

All 112 in Engineering and CAD
  • EEBench In use 13 electrical-engineering circuit and PCB design tasks graded by deterministic build checks, simulation and BOM constraints. No LLM judge. Gemini 4 Carbon 83.9%
  • 3DCodeBench In use Agentic procedural 3D modeling via code: author engine-ready 3D assets through software APIs. Graded for executability and shape fidelity. LongCat-2.5 55.4%
  • CADGenBench Only in system cards Agents construct CAD models from design prompts. The generation split is graded for executability and geometric correctness. Claude Fable 5 45.1%
  • PartBench Only in system cards Parametric mechanical part design and CAD-style generation, graded for geometric and manufacturing fidelity. Grok 4.6 58.5%
  • Razavi-Bench Not in model cards Models explain what happens in analog circuits and why. A language model awards up to four points per answer against a reference and a rubric. Gemini 4 Carbon 98.3%
  • CircuitRubric Not in model cards Models write analog circuit netlists from design requests. A program compares transistor connections and relative device sizes with accepted reference circuits. Claude Opus 4.8 96%

Data audit

All 8 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
  • Pass Independent runs Scores from the official leaderboard (24 entries).
  • Pass Models scored 24 entries on its leaderboard, none a model we track.
  • Note Newest score No scores yet.
  • Note Frontier models No models scored yet.
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on xilinx.github.io: 24 entries at the last read, none a model we track.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check