Circuit design for analog integrated circuits (ICs) is challenging because circuit topology, device sizing, and circuit performance are tightly coupled. Existing large language model (LLM)-based methods either rely on inference-time correction around frozen models or train on limited circuit types without directly using simulation outcomes as training signals. A key bottleneck is that existing datasets do not provide sufficient task information for training LLMs on end-to-end specification-conditioned design. To fill this gap, we first construct a large-scale simulation-grounded dataset with 8,626 design tasks across 10 different types of analog circuits, where each task includes a textual description, labeled I/O ports, loading condition, target specifications, a testbench, and a reference netlist. Using this dataset, we present SAFTAC, a simulation-augmented fine-tuning framework that trains LLMs to generate and revise analog circuit design. SAFTAC combines three-stage supervised fine-tuning, which progressively builds the capabilities required for analog circuit design and feedback-guided correction, with simulation feedback-augmented reinforcement fine-tuning, which further optimizes models using simulation-grounded rewards. We evaluate SAFTAC under single-pass generation and two-pass generation with simulation feedback, showing improvements over the strongest closed-source LLM by 15.9% and 7.8%, respectively.
Figures & tables
Figure 1: Overview of the automated data generation pipeline for analog circuit design tasks. (a) Raw circuit sources are collected and transformed into netlists with circuit-type labels, then augmented with textual descriptions, circuit load, testbenches, and Pareto-feasible specifications derived through BO sizing and SPICE simulation. Each step is guided and checked by human experts. (b) A data sample is provided for reference, with the Python testbench code omitted for brevity.
Feature
Analogcoder-Pro Lai et al. (2026)
AnalogGym Li et al. (2024)
AMSNet-KG Shi et al. (2025)
AnalogGenie Gao et al. (2025)
Masala-CHAI Bhandari et al. (2024)
Ours
Text Descriptions
✓
✗
✓
✗
✓
✓
Multiple Specs
✓
✓
✓
✗
✗
✓
Labeled I/O
✓
✓
✓
✗
✗
✓
Testbench
✓
✓
✓
✗
✗
✓
Circuit Load
✗
✓
✓
✗
✗
✓
Open-Source
✓
✓
✗
✓
✓
✓
Table 1: Comparison with existing analog circuit design datasets. We compare both task components and dataset scale. Our dataset provides the most complete set of task components while covering 10 circuit types and 8,626 design tasks. Masala-CHAI Bhandari et al. (2024) does not disclose circuit-type information, so its number of circuit types and average circuits per type are marked as unavailable.
Method
Text Specs
Opt. Sizing
Sim.-in-Training
Training
Multi-Type
Open-Source
Analogcoder Lai et al. (2025)
✓
✗
–
–
✓
▲
Analogcoder-Pro Lai et al. (2026)
✓
✓
–
–
✓
▲
AnalogSAGE Wang et al. (2025)
✓
✓
–
–
✗
▲
AnalogGenie Gao et al. (2025)
✗
✗
✗
Pre-Training
✓
✓
LaMAGIC Chang et al. (2024)
✗
✗
✗
SFT
✗
✗
Artisan Chen et al. (2024)
✓
✓
✗
SFT
✗
✗
Table 2: Comparison with LLM-based analog circuit design methods. We compare whether each method includes textual specifications, optimized sizing, incorporates simulation in the training stage, trains the underlying model, supports multiple circuit types, and is open-source. Our method is the only one that supports all these dimensions. ▲ indicates that only the code is released as open source.
SFT stage
Task Summary
#
Circuit Knowledge Acquisition
Describe the circuit topology from a complete netlist.
2,949
Masked Component Prediction
Predict the missing component or connection in an incomplete netlist.
1,967
Simulation Feedback-Guided Correction
Revise a netlist using prior errors and simulation feedback.
921
Table 3: Overview of the three-stage SFT. # denotes the number of training examples used.
Figure 2: Overview of the simulation feedback-augmented RFT pipeline. It combines standard circuit design tasks and simulation feedback-guided correction tasks. In both tasks, the generated netlist is evaluated through circuit topology check, BO sizing, and simulation. The resulting outcome is converted into a reward signal and used to update the policy with GRPO.
Model Type
Model
Common Gate
Common Source
Current Mirror
Source Follower
Passive Mixer
Average
Closed-source
GPT-5.2
50.4
65.6
61.3
58.2
26.0
52.3
GPT-5.4
48.0
65.6
58.0
58.2
29.1
51.8
Claude-Opus-4.7
45.7
65.6
62.7
63.7
39.8
55.5
Open-source
QwQ-32B
41.0
63.5
27.9
35.2
25.6
38.6
Qwen3-32B
42.0
55.3
38.3
34.9
29.3
40.0
gpt-oss-20b
42.2
57.2
7.8
34.6
16.8
31.7
Table 4: Pass rate comparison (%) under single-pass generation. The model directly generates a complete circuit netlist from the design specification. We report pass@1 over 8 rollouts per test instance. Bold indicates the best result in each column.
Model Type
Model
Common Gate
Common Source
Current Mirror
Source Follower
Passive Mixer
Average
Closed-source
GPT-5.2
59.8
75.8
81.6
68.6
37.7
64.7
GPT-5.4
66.4
78.1
78.7
69.7
41.8
66.9
Claude-Opus-4.7
63.1
76.6
78.7
79.5
54.1
70.4
Open-source
QwQ-32B
45.5
68.0
30.1
50.4
26.6
44.1
Qwen3-32B
52.9
71.7
58.0
50.7
33.2
53.3
gpt-oss-20b
49.2
68.8
29.7
39.1
18.8
41.1
Table 5: Pass rate (%) comparison under two-pass generation with simulation feedback. The model first generates an initial circuit design and may revise it once using simulation-guided feedback. We report the overall pass@1 after up to two generation attempts, where a rollout is counted as successful if either the initial design or the revised design satisfies all target specifications. Bold indicates the best result in each column.
Figure 3: Ablation study on different training components in SAFTAC . We compare different model variants under single-pass generation and two-pass generation with simulation feedback. The correction rate measures the fraction of initially failed designs that are successfully repaired in the second pass. Full SAFTAC achieves the best performance across all three metrics.
Setting
Reward Design
Common Gate
Common Source
Current Mirror
Source Follower
Passive Mixer
Average
Single-pass
Partial Reward
46.5
70.7
61.3
68.4
62.1
61.8
Original Reward
56.4
79.5
75.8
77.5
67.6
71.4
Two-pass
Partial Reward
57.8
76.6
72.5
73.4
75.6
71.2
Original Reward
67.8
81.8
82.8
80.1
78.5
78.2
Table 6: Ablation study on reward design. We compare the original reward with an alternative reward based on partial specification satisfaction. We report pass@1 (%) under single-pass generation and two-pass generation with simulation feedback. Bold indicates the best result in each column.
Backbone Model
Single-pass Generation
Two-pass Generation with Simulation Feedback
Base
SAFTAC
Base
SAFTAC
Qwen3-8B
27.4
71.4
28.7
78.2
Qwen3-4B
25.4
46.0
30.6
57.7
Llama-3.1-8B-Instruct
23.6
77.5
28.7
83.6
GLM-4-9B
26.5
54.5
29.6
68.8
Table 7: Generalizability of SAFTAC across backbone models. We apply the complete training pipeline to open-source LLMs from different model families and scales. We report the average pass@1 (%) across five circuit types under single-pass generation and two-pass generation with simulation feedback. Bold indicates the results after applying SAFTAC.
Model
Active Mixer
Differential Single-Ended Amplifier
Fully Differential Amplifier
Inverting Delay Cell
Non-Inverting Delay Cell
pass@1
pass@5
pass@1
pass@5
pass@1
pass@5
pass@1
pass@5
pass@1
pass@5
gpt-oss-20b
4.7
20.1
2.7
12.6
44.1
80.8
8.6
35.1
0.0
0.0
Qwen3-8B
1.6
7.8
0.0
0.0
18.4
37.7
11.7
39.1
0.0
0.0
SAFTAC (Ours)
41.4
85.0
3.1
12.6
58.6
88.1
16.4
43.0
6.3
31.3
Table 8: OOD evaluation results on unseen circuit types. We evaluate models on circuit types excluded from training and report pass@1 and pass@5 under two-pass generation with simulation feedback . We sample 8 rollouts for each task. Bold indicates the best result in each column.
Base Model
Amplifier
Inverter
Current Mirror
Analog Adder
pass@1
pass@5
pass@1
pass@5
pass@1
pass@5
pass@1
pass@5
GPT-4o w/ AnalogCoder
100.0
100.0
100.0
100.0
56.7
78.6
100.0
100.0
GPT-5.4 w/ AnalogCoder
100.0
100.0
100.0
100.0
56.7
78.6
93.3
100.0
Qwen3-32B w/ AnalogCoder
99.1
100.0
100.0
100.0
26.7
65.7
73.3
100.0
QwQ-32B w/ AnalogCoder
60.0
99.0
100.0
100.0
60.0
95.8
93.3
100.0
SAFTAC w/o AnalogCoder
98.1
100.0
100.0
100.0
76.7
100.0
66.7
100.0
Table 9: Compatibility with the AnalogCoder inference pipeline. We evaluate different base models within the AnalogCoder Lai et al. (2025) inference-time pipeline and report pass@1/pass@5 across circuit types. SAFTAC w/o AnalogCoder denotes direct generation without the AnalogCoder pipeline.
Figure 4: Simulation-guided circuit revision examples. The top row shows initial failed designs, and the bottom row shows the corresponding revisions after incorporating simulation-augmented feedback. Cases (a) and (b) show structural recovery from topology-check failures, while cases (c) and (d) show metric-driven adjustment for specification violations. Red marks removed parts, and green marks newly introduced parts. Detailed information is provided in App. C.2 .
Appendix figures & tables5 assets
Supplementary material from the paper’s appendix.
Appendix
Circuit Type
Performance Specs
# Unique Reference Netlists
# Design Tasks
Common Gate Amplifier
gain , bandwidth , power
15
1,372
Common Source Amplifier
gain , bandwidth , power
18
1,861
Source Follower
gain , bandwidth , power
8
720
Current Mirror
rout , Vout,max , Vout,min , current ratio
12
478
Differential Single-Ended Amplifier
gain , bandwidth , output swing , phase margin , CMRR , power
23
955
Fully Differential Amplifier
gain , bandwidth , output swing , phase margin , CMRR , power
11
1,632
Appendix
Table 10: Dataset statistics by circuit type. We report the number of unique reference netlists and specification-conditioned design tasks for each circuit type. Multiple design tasks can be derived from the same reference netlist with different target specifications. Performance specifications are color-coded by optimization direction: red indicates higher-is-better, green indicates lower-is-better, and black indicates equality constraints. See App. B.2 for specification definitions.
Knowledge Acquisition
Component Prediction
Simulation Feedback-Guided Correction
Epochs
1
1
1
Total Effective Batch Size
64
64
64
Warm-up Ratio
0.00
0.00
0.00
Learning Rate
1e-5
1e-5
1e-5
GPU Hardware
8x NVIDIA RTX PRO 6000
Appendix
Table 11: Detailed three-stage SFT recipes.
Optimizer
AdamW
Total Batch Size
640
Mini Batch Size
320
Learning Rate
1e-6
LR Schedule
Constant
Weight Decay
0
Warm-up Ratio
0
Appendix
Table 12: Detailed RFT recipes.
Setting
Model
Invalid Syntax
Topology Failure
BO Sizing Failure
Spec. Not Met
Success
Single-pass
Claude-Opus-4.7
0.1
6.5
0.1
37.8
55.5
GPT-5.4
0.0
7.9
0.0
40.3
51.8
QwQ-32B
25.7
4.6
0.0
31.1
38.6
Qwen3-32B
22.3
3.6
0.0
34.1
40.0
gpt-oss-20b
32.3
4.4
0.0
31.6
31.7
SAFTAC (Ours)
0.0
4.2
1.1
23.3
71.4
Appendix
Table 13: Failure mode distribution (%) across different models. Generated circuits are categorized into five mutually exclusive outcomes based on deterministic circuit validation. Results are reported for single-pass generation and two-pass generation with simulation feedback.
Model
Unique Successful Topologies
Device Count Single-pass
Device Count Two-pass
Claude-Opus-4.7
32
3.59
4.09
GPT-5.4
59
4.70
5.06
SAFTAC (Ours)
81
3.54
3.50
Appendix
Table 14: Topology diversity and correction behavior. We report the number of unique successful circuit topologies generated across five circuit types, where a higher count indicates greater observed diversity under the same evaluation budget. Average device counts before and after correction characterize circuit size. Bold indicates the largest number of unique successful topologies.