LFM2-8B-A1B-mic

Slim model weights, executable usage and framework expectations in one repository.

• Contract manifest
• Transformers usage
• Transformers expectations
• Fixture provenance
• How the fixture was shrunk

Contract

LFM2-8B-A1B-mic

Slim model weights, executable usage and framework expectations in one repository.

• Contract manifest
• Transformers usage
• Transformers expectations
• Fixture provenance
• How the fixture was shrunk

Contract

LFM2-8B-A1B causal LM

Runnable usage and the framework expectations directory are linked above.

Slim model

Slim fixture: LFM2-8B-A1B-mic

A 4.4M-parameter fixture for the LFM2-8B-A1B MIC, covering LiquidAI/LFM2-8B-A1B. It is a test artifact, not a usable model, tagged test-artifact to keep it out of candidate-model search.

• Fixture key fac5a8d2dc0e. It covers every checkpoint whose processor files are byte-identical and whose config matches after shrinking.
• Same architecture, smaller config: Lfm2MoeForCausalLM, shrunk by the generic rule in sweep.py (num_hidden_layers 4).
• Processor files copied byte for byte from LiquidAI/LFM2-8B-A1B: LICENSE, chat_template.jinja, generation_config.json, special_tokens_map.json, tokenizer.json, tokenizer_config.json.
• Random weights (seed 17). They passed the input-sensitivity gate: every perturbation of the contract's inputs moved the logits by at least 0.001 (weakest: attention_mask:all_ones, 0.258), so no training was needed.
• Same tensor names as the real checkpoint. The real checkpoint has every parameter, and so does the fixture.
• Saved in bfloat16, the dtype LiquidAI/LFM2-8B-A1B loads in by default.
• Finite in float32, bfloat16, and float16 (forward, loss, gradients).

Built by build_fixtures.py; build.json records the pinned sources, gate results, and file hashes. This repository includes the slim weights, usage and framework expectations; rebuild the other files with python example/build_fixtures.py --fixture LFM2-8B-A1B-mic.

Slim model

LFM2-8B-A1B-mic

Slim model weights, executable usage and framework expectations in one repository.

• Contract manifest
• Transformers usage
• Transformers expectations
• Fixture provenance

Contract

LFM2-8B-A1B causal LM

Runnable usage and the framework expectations directory are linked above.

Slim model

Slim fixture: LFM2-8B-A1B-mic

A 4.4M-parameter fixture for the LFM2-8B-A1B MIC, covering LiquidAI/LFM2-8B-A1B. It is a test artifact, not a usable model, tagged test-artifact to keep it out of candidate-model search.

• Fixture key fac5a8d2dc0e. It covers every checkpoint whose processor files are byte-identical and whose config matches after shrinking.
• Same architecture, smaller config: Lfm2MoeForCausalLM, shrunk by the generic rule in sweep.py (num_hidden_layers 4).
• Processor files copied byte for byte from LiquidAI/LFM2-8B-A1B: LICENSE, chat_template.jinja, generation_config.json, special_tokens_map.json, tokenizer.json, tokenizer_config.json.
• Random weights (seed 17). They passed the input-sensitivity gate: every perturbation of the contract's inputs moved the logits by at least 0.001 (weakest: attention_mask:all_ones, 0.258), so no training was needed.
• Same tensor names as the real checkpoint. The real checkpoint has every parameter, and so does the fixture.
• Saved in bfloat16, the dtype LiquidAI/LFM2-8B-A1B loads in by default.
• Finite in float32, bfloat16, and float16 (forward, loss, gradients).

Built by build_fixtures.py; build.json records the pinned sources, gate results, and file hashes. This repository includes the slim weights, usage and framework expectations; rebuild the other files with python example/build_fixtures.py --fixture LFM2-8B-A1B-mic.