Instructions to use veriloop-lab/veriloop-coder-e1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use veriloop-lab/veriloop-coder-e1 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="veriloop-lab/veriloop-coder-e1")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("veriloop-lab/veriloop-coder-e1")
model = AutoModelForImageTextToText.from_pretrained("veriloop-lab/veriloop-coder-e1")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

PEFT
How to use veriloop-lab/veriloop-coder-e1 with PEFT:
```
Task type is invalid.
```
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use veriloop-lab/veriloop-coder-e1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "veriloop-lab/veriloop-coder-e1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "veriloop-lab/veriloop-coder-e1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/veriloop-lab/veriloop-coder-e1

SGLang

How to use veriloop-lab/veriloop-coder-e1 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "veriloop-lab/veriloop-coder-e1" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "veriloop-lab/veriloop-coder-e1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "veriloop-lab/veriloop-coder-e1" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "veriloop-lab/veriloop-coder-e1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use veriloop-lab/veriloop-coder-e1 with Docker Model Runner:
```
docker model run hf.co/veriloop-lab/veriloop-coder-e1
```

veriloop-coder-e1 / rollback_adapter /rollback_adapter_train_result.json

ConorWang

Upload rollback adapter artifacts to rollback_adapter/

0a1c0ab verified about 1 month ago

raw

history blame

14.1 kB

	{
	"artifacts": {
	"adapter_dir": "/private/wang_libo/veriloop_coder_e1/outputs/rollback_qwen36_rootfix_run1/adapter",
	"best_checkpoint_manifest": "/private/wang_libo/veriloop_coder_e1/outputs/rollback_qwen36_rootfix_run1/best_checkpoint_manifest.json",
	"epoch_history": "/private/wang_libo/veriloop_coder_e1/outputs/rollback_qwen36_rootfix_run1/epoch_history.json",
	"eval_jsonl": "/private/wang_libo/veriloop_coder_e1/outputs/rollback_qwen36_rootfix_run1/rollback_eval.jsonl",
	"host_manifest": "/private/wang_libo/veriloop_coder_e1/outputs/rollback_qwen36_rootfix_run1/host_manifest.json",
	"plan_json": "/private/wang_libo/veriloop_coder_e1/outputs/rollback_qwen36_rootfix_run1/rollback_adapter_plan.json",
	"rollback_head": "/private/wang_libo/veriloop_coder_e1/outputs/rollback_qwen36_rootfix_run1/rollback_head.pt",
	"tokenizer_dir": "/private/wang_libo/veriloop_coder_e1/outputs/rollback_qwen36_rootfix_run1/tokenizer",
	"train_jsonl": "/private/wang_libo/veriloop_coder_e1/outputs/rollback_qwen36_rootfix_run1/rollback_train.jsonl",
	"training_manifest": "/private/wang_libo/veriloop_coder_e1/outputs/rollback_qwen36_rootfix_run1/rollback_training_manifest.json"
	},
	"dataset": {
	"cause_vocab": [
	"none",
	"evidence_contradiction",
	"execution_negation",
	"validator_rejection",
	"patch_regression",
	"worktree_conflict",
	"high_risk_unsupported",
	"tool_selfcheck_failure",
	"reverse_engineering_speculation"
	],
	"eval_size": 45,
	"mode_vocab": [
	"contradiction",
	"execution_failure",
	"validator_rejection",
	"patch_regression",
	"worktree_conflict",
	"high_risk_unsupported",
	"tool_selfcheck_repair",
	"reverse_engineering_drift",
	"validated_stable"
	],
	"modes": [
	"contradiction",
	"execution_failure",
	"high_risk_unsupported",
	"patch_regression",
	"reverse_engineering_drift",
	"tool_selfcheck_repair",
	"validated_stable",
	"validator_rejection",
	"worktree_conflict"
	],
	"need_vocab": [
	"none",
	"advisory",
	"required"
	],
	"rewrite_vocab": [
	"do_not_rollback",
	"local_patch_only",
	"validator_scoped_revision",
	"minimum_diff_patch",
	"do_not_overwrite_worktree_state",
	"fail_closed",
	"repair_via_bounded_selfcheck_loop",
	"bounded_observation_first",
	"local_conflict_preserving_revision"
	],
	"train_size": 126
	},
	"eval_metrics": {
	"adapter_exported": true,
	"auto_lora_from_ia3": false,
	"avg_binary_accuracy": 0.8194444444444444,
	"best_epoch": 3,
	"best_quality_score": 0.3986773560841878,
	"cause_accuracy": 0.4888888888888889,
	"count": 45,
	"eval_batches": 45,
	"eval_loss": 7.656409973568387,
	"fail_closed_accuracy": 0.7777777777777778,
	"minimal_diff_accuracy": 0.5555555555555556,
	"mode_accuracy": 0.3333333333333333,
	"need_accuracy": 0.7777777777777778,
	"peft_method": "lora_narrow",
	"preserve_conflict_accuracy": 0.7777777777777778,
	"quality_score": 0.3986773560841878,
	"reverse_engineering_bounded_accuracy": 0.8888888888888888,
	"rewrite_accuracy": 0.4888888888888889,
	"selfcheck_route_accuracy": 0.5555555555555556,
	"session_continuity_accuracy": 1.0,
	"used_peft": true,
	"validator_aligned_accuracy": 1.0,
	"worktree_safe_accuracy": 1.0
	},
	"plan": {
	"backbone": "/public/wang_libo/veriloop_coder_e1/model",
	"dataset_summary": {
	"cause_vocab": [
	"none",
	"evidence_contradiction",
	"execution_negation",
	"validator_rejection",
	"patch_regression",
	"worktree_conflict",
	"high_risk_unsupported",
	"tool_selfcheck_failure",
	"reverse_engineering_speculation"
	],
	"eval_size": 45,
	"mode_vocab": [
	"contradiction",
	"execution_failure",
	"validator_rejection",
	"patch_regression",
	"worktree_conflict",
	"high_risk_unsupported",
	"tool_selfcheck_repair",
	"reverse_engineering_drift",
	"validated_stable"
	],
	"modes": [
	"contradiction",
	"execution_failure",
	"high_risk_unsupported",
	"patch_regression",
	"reverse_engineering_drift",
	"tool_selfcheck_repair",
	"validated_stable",
	"validator_rejection",
	"worktree_conflict"
	],
	"need_vocab": [
	"none",
	"advisory",
	"required"
	],
	"rewrite_vocab": [
	"do_not_rollback",
	"local_patch_only",
	"validator_scoped_revision",
	"minimum_diff_patch",
	"do_not_overwrite_worktree_state",
	"fail_closed",
	"repair_via_bounded_selfcheck_loop",
	"bounded_observation_first",
	"local_conflict_preserving_revision"
	],
	"train_size": 126
	},
	"excluded_surfaces": [
	"(^\|\\.)lm_head($\|\\.)::Do not retune final token head; too broad and evaluation-heavy.",
	"(^\|\\.)embed_tokens($\|\\.)::Embedding surgery risks broad semantic drift.",
	"(^\|\\.)norm($\|\\.)::Global norm tuning can destabilize calibration across scenes.",
	"attnres\|attention_residual::Block AttnRes may be mounted structurally but is never a PEFT target.",
	"dualpath::DualPath is serving/runtime infrastructure only.",
	"mhc\|hyper[-_]?connection::mHC-inspired stability hooks remain structural, not PEFT surfaces.",
	"rope\|rotary::RoPE/context surgery is handled architecturally, not by narrow PEFT here.",
	"kvcache\|kv_cache::KV-cache runtime surfaces are not PEFT targets.",
	"(^\|\\.)memory(_store\|_bank)?($\|\\.)::Persistent memory stores are harness/runtime policy surfaces, not PEFT targets."
	],
	"notes": [
	"Primary route is host-surface-first rollback training.",
	"Validator receipts, failure routing, patch state, worktree continuity, self-check repair, and reverse-engineering boundedness are first-class signals.",
	"DualPath, Block AttnRes, mHC hooks, and MoE routers/experts remain structurally excluded.",
	"This adapter should improve rollback obedience, not general coding free-formity."
	],
	"peft_method": "lora_narrow",
	"product_line": "veriloop_coder",
	"recipe": {
	"adapter_family": "validator_rollback",
	"backbone": "/public/wang_libo/veriloop_coder_e1/model",
	"backbone_family": "qwen_dense",
	"excluded_patterns": [
	"(?i)\\bdualpath\\b",
	"(?i)\\bmhc\\b",
	"(?i)\\bfull[_\\- ]?attnres\\b",
	"(?i)\\battnres(_full)?\\b",
	"(?i)\\brouter\\b",
	"(?i)\\bexperts?\\b",
	"(?i)\\bmoe\\b.*\\b(gate\|router\|expert)\\b",
	"(?i)\\brope\\b.*\\b(freq\|inv_freq\|theta\|rotary)\\b",
	"(?i)\\bkvcache\\b",
	"(?i)\\bposition_embedding\\b",
	"(?i)\\bembed(tokens\|ding)?\\b",
	"(?i)\\blm_head\\b"
	],
	"harness_constraints": [
	"Harness Engineering remains the primary convergence layer.",
	"Adapter must not bypass runtime orchestrator / validator / rollback loops.",
	"Adapter outputs remain subordinate to VeriLoop control-plane decisions.",
	"Adapter must not create hidden prompt-style memory authority.",
	"Adapter must preserve validator-first correction and minimal-diff rollback discipline.",
	"Patch production must remain subordinate to validator receipts and failure routing."
	],
	"hyperparams": {
	"alpha": 16,
	"bias": "none",
	"dropout": 0.08,
	"fan_in_fan_out": false,
	"modules_to_save": [
	"input_layernorm"
	],
	"r": 8,
	"task_type": "CAUSAL_LM"
	},
	"merge_policy": "side_load",
	"metadata": {
	"allow_backbone_bridge": false,
	"allow_vla_action_expert": false,
	"harness_first": true,
	"prefer_explicit_heads": true,
	"prefer_qlora_for_backbone_bridge": false,
	"require_harness_first": true,
	"reverse_engineering_readiness": true,
	"rollback_training": true,
	"selector_group_count": 2,
	"selfcheck_loop_readiness": true,
	"strict_narrow_scope": true,
	"trainer": "veriloop.rollback_adapter_trainer.v9.qwen36"
	},
	"notes": [
	"Backbone bridge tuning disabled explicitly; selector stays on custom surfaces or no-op.",
	"Backbone family inferred as qwen_dense.",
	"PEFT method resolved as lora_narrow.",
	"Recipe is harness-first: runtime convergence remains in VeriLoop control-plane + harness, not in broad weight surgery.",
	"Block AttnRes, DualPath, mHC hooks, RoPE, KV-cache, and broad MoE routing remain structurally excluded."
	],
	"peft_method": "lora_narrow",
	"precision_policy": "auto",
	"product_line": "veriloop_coder",
	"regression_requirements": [
	"Must pass PEFT regression guard structural policy checks.",
	"Must not introduce forbidden backbone/serving structural targets.",
	"Must preserve harness regression envelope for the selected product line.",
	"Validator feedback uptake must not regress.",
	"Rollback minimality and revision fidelity must not regress."
	],
	"target_groups": [
	{
	"alpha": 16,
	"dropout": 0.05,
	"name": "group_1_custom_runtime_harness_bridge",
	"rank": 8,
	"rationale": "Runtime / harness obedience should attach to explicit interface bridges before any backbone fallback.",
	"surface": "custom_runtime_harness_bridge",
	"target_modules": [
	"failure_signal_bridge.rollback_bridge",
	"request_normalizer",
	"request_normalizer.adapter",
	"rollback_adapter",
	"rollback_adapter.head",
	"runtime_harness_adapter",
	"runtime_harness_adapter.bridge",
	"tool_protocol_adapter",
	"tool_protocol_adapter.bridge",
	"toolspec_bridge",
	"toolspec_bridge.adapter",
	"toolspec_head",
	"toolspec_head.param_schema_adapter",
	"toolspec_head.postcondition_adapter",
	"toolspec_head.precondition_adapter",
	"toolspec_head.receipt_formatter",
	"toolspec_head.trigger_gate",
	"toolspec_head.validator_gate",
	"validator_feedback_loop.rollback_adapter"
	]
	},
	{
	"alpha": 16,
	"dropout": 0.0,
	"name": "group_2_custom_validator_bridge",
	"rank": 8,
	"rationale": "Validation and rollback fidelity should prefer explicit validator / rollback bridges.",
	"surface": "custom_validator_bridge",
	"target_modules": [
	"failure_signal_bridge",
	"failure_signal_bridge.rollback_bridge",
	"rollback_adapter",
	"rollback_adapter.head",
	"sandbox_rollback_bridge",
	"sandbox_rollback_bridge.adapter",
	"validator_feedback_bridge",
	"validator_feedback_bridge.adapter",
	"validator_feedback_loop.rollback_adapter"
	]
	}
	],
	"target_modules": [
	"failure_signal_bridge.rollback_bridge",
	"request_normalizer",
	"request_normalizer.adapter",
	"rollback_adapter",
	"rollback_adapter.head",
	"runtime_harness_adapter",
	"runtime_harness_adapter.bridge",
	"tool_protocol_adapter",
	"tool_protocol_adapter.bridge",
	"toolspec_bridge",
	"toolspec_bridge.adapter",
	"toolspec_head",
	"toolspec_head.param_schema_adapter",
	"toolspec_head.postcondition_adapter",
	"toolspec_head.precondition_adapter",
	"toolspec_head.receipt_formatter",
	"toolspec_head.trigger_gate",
	"toolspec_head.validator_gate",
	"validator_feedback_loop.rollback_adapter",
	"failure_signal_bridge",
	"sandbox_rollback_bridge",
	"sandbox_rollback_bridge.adapter",
	"validator_feedback_bridge",
	"validator_feedback_bridge.adapter"
	],
	"version": "veriloop.lora_recipe_veriloop.v2"
	},
	"selected_surfaces": [
	"custom_runtime_harness_bridge",
	"custom_validator_bridge",
	"custom_memory_boundary_bridge"
	],
	"selected_target_modules": [
	"failure_signal_bridge.rollback_bridge",
	"rollback_adapter.head",
	"runtime_harness_adapter.bridge",
	"sandbox_rollback_bridge.adapter",
	"toolspec_head.validator_gate",
	"validator_feedback_bridge.adapter",
	"validator_feedback_loop.rollback_adapter",
	"memory_boundary_guard.adapter",
	"memory_boundary_guard.rollback_filter",
	"request_normalizer.adapter",
	"tool_protocol_adapter.bridge",
	"toolspec_bridge.adapter",
	"toolspec_head.postcondition_adapter",
	"toolspec_head.precondition_adapter",
	"toolspec_head.receipt_formatter",
	"toolspec_head.trigger_gate"
	],
	"selection_mode": "minimal",
	"version": "veriloop.rollback_adapter_trainer.v9.qwen36",
	"warnings": [
	"Harness Engineering is primary; PEFT is limited to obedience-facing, interface-facing support surfaces.",
	"Backbone bridge tuning disabled explicitly; selector stays on custom surfaces or no-op."
	]
	},
	"status": "trained",
	"train_metrics": {
	"adapter_exported": true,
	"auto_lora_from_ia3": false,
	"best_epoch": 3,
	"best_quality_score": 0.3986773560841878,
	"epochs_completed": 4,
	"loss": 0.4452953128114579,
	"micro_batches": 126,
	"micro_batches_total": 504,
	"optimizer_steps": 8,
	"optimizer_steps_total": 32,
	"peft_method": "lora_narrow",
	"used_peft": true
	},
	"version": "veriloop.rollback_adapter_trainer.v9.qwen36",
	"warnings": [
	"Harness Engineering is primary; PEFT is limited to obedience-facing, interface-facing support surfaces.",
	"Backbone bridge tuning disabled explicitly; selector stays on custom surfaces or no-op."
	]
	}