ironbench
Agents write firmware for tasks such as serial communication and controller sequencing. The grader checks device behavior and counts tasks solved within the allowed attempts.
The September 10 archive compares five models across 13 tasks with two attempts each. Later checks found three control tasks could accept arbitrary output.
- Made by
- Tsyren Balmaev
- Released
- September 2026
- Kind
- Electronics
- Level
- University
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Agentic
- Environment
- A simulator, APIs and tools
- Refresh
- One fixed set
- Run by
- Anyone
- Strongest model tested
-
Qwen3.5-9B #128 on the capability index
- Last model added
- February 2026
- Scores refreshed
- November 20th, 2026
glm-4.7-flash leads with 69.2% among models released by February 2026 glm-4.7-flash leads with 69.2% among models released by February 2026
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. This view shows only models we track, each at its best.
The newest model with a score here came out in February 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 1 glm-4.7-flash 69.2%
- 2 GLM-4.6V-Flash 53.9%
- 3 qwen2.5-7b-instruct 53.9%
- 4
Qwen3.5-4B 46.1%
- 5
Qwen3.5-9B 38.5%
Source: Official ironbench leaderboard
- 1 glm-4.7-flash 69.2%
- 2 GLM-4.6V-Flash 53.9%
- 3
Qwen3.5-4B 46.1%
- 4
Qwen3.5-9B 38.5%
Source: Official ironbench leaderboard
| # | Entry | Score | Setting | Released | Run on | Source |
|---|---|---|---|---|---|---|
| 1 | glm-4.7-flash | 69.2% | Jan 2026 | Sep 2026 | Official leaderboard | |
| 2 | GLM-4.6V-Flash | 53.9% | Dec 2025 | Sep 2026 | Official leaderboard | |
| 3 | qwen2.5-7b-instruct | 53.9% | Sep 2026 | Official leaderboard | ||
| 4 | | 46.1% | Feb 2026 | Sep 2026 | Official leaderboard | |
| 5 | | 38.5% | Feb 2026 | Sep 2026 | Official leaderboard |
| # | Model | Score | Setting | Released | Run on | Source |
|---|---|---|---|---|---|---|
| 1 | glm-4.7-flash | 69.2% | Jan 2026 | Sep 2026 | Official leaderboard | |
| 2 | GLM-4.6V-Flash | 53.9% | Dec 2025 | Sep 2026 | Official leaderboard | |
| 3 | | 46.1% | Feb 2026 | Sep 2026 | Official leaderboard | |
| 4 | | 38.5% | Feb 2026 | Sep 2026 | Official leaderboard |
Rankings by part
We read 5 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Control, FSM and IO. glm-4.7-flash leads 3 of them, and GLM-4.6V-Flash leads the other 2. Arrows show how far a model moved from its overall rank among the models we track.
Control
4 models we track have a score here.
- 1 glm-4.7-flash 75%
- 2 GLM-4.6V-Flash 50%
- 3
Qwen3.5-9B ↑1 37.5%
- 4
Qwen3.5-4B ↓1 12.5%
FSM
4 models we track have a score here.
- 1 GLM-4.6V-Flash ↑1 50%
- 2 glm-4.7-flash ↓1 50%
- 3
Qwen3.5-4B 50%
- 4
Qwen3.5-9B 0%
IO
4 models we track have a score here.
- 1 GLM-4.6V-Flash ↑1 100%
- 2 glm-4.7-flash ↓1 100%
- 3
Qwen3.5-4B 100%
- 4
Qwen3.5-9B 100%
Protocol
4 models we track have a score here.
- 1 glm-4.7-flash 25%
- 2
Qwen3.5-4B ↑1 12.5%
- 3 GLM-4.6V-Flash ↓1 0%
- 4
Qwen3.5-9B 0%
Resilience
4 models we track have a score here.
- 1 glm-4.7-flash 50%
- 2
Qwen3.5-4B ↑1 37.5%
- 3 GLM-4.6V-Flash ↓1 25%
- 4
Qwen3.5-9B 12.5%
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official ironbench leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official ironbench 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
- 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
9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 9 months ago, in February 2026, for Qwen3.5-4B. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
- Pass Independent runs Scores from the official leaderboard (5 entries).
- Pass Models scored 4 models we track, plus 1 other entry on its leaderboard.
- Warning Newest score The newest came 9 months ago, in February 2026, for Qwen3.5-4B.
- Note Frontier models None of the top 10 has a score. The strongest tested is Qwen3.5-9B (#128).
- Pass Headroom The best is 69.2%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on cezman.github.io: 4 rows for models we track and 1 other entry at the last read.
- 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.