Lumma-Fev-27B has 26.085 billion total parameters. The selected Q4_K_M GGUF file is 16.547 GB.
Data snapshot: . 1 published test in this catalog.
Model file and details
Provider
FrontiersMind
Total parameters
26,085,330,160
Exact file size
16,547,401,888 bytes
Quantization / precision
Q4_K_M
Weight repository
FrontiersMind/Lumma-fev-27B-GGUF
Original model repository
FrontiersMind/Lumma-fev-27B
Weight revision
c420dca09983849cdd21d42c1b93d89d99482755
Model revision
2c080fb4b44bbad2a140353393d935e8b7af8bcb
Runtime notes
Confirm model architecture and quantization support in your runner.
Recorded release date
2026-10-14
Date evidence
FrontiersMind released the Apache-2.0 source checkpoint on 13 October 2026 and a same-publisher GGUF repository on 14 October. The selected 16,547,401,888-byte Q4_K_M language file is a compact catalog artifact, but the reported decision accuracy belongs to the 26,085,330,160-parameter source checkpoint and was not transferred to the quantized file. Lumma-Fev-27B is a typed decision model for English, Hindi, Tamil and Kannada; it returns option probabilities rather than generated chat text.
These scores were compiled from the linked sources. These published scores were not independently tested here. Different prompts, precision, thinking budgets and tools can change results.
Lumma-Fev-27B · held-out decisions
90.7 / 100
Publisher-reported source-checkpoint accuracy on 4,454 held-out typed decisions from nine open task families. This is a specialist decision test, not a general chat benchmark or a quantized-artifact result.
Benchmark efficiency: 5.481 published score points per GB of selected weights. This ratio is not an inference-speed measure; check the test settings to see whether the score covers the selected quantization.
Test settings: Publisher-run source-checkpoint evaluation in October 2026 on 4,454 held-out noul and choice decisions from nine open task families. Lumma-Fev-27B and hosted TypeSafe Jev 1.13 were evaluated on the same rows; the publisher reports 90.7% versus 83.9% accuracy and 0.009 versus 0.045 expected calibration error.
Source-checkpoint result, not an independent evaluation and not a measurement of the selected Q4_K_M GGUF. The publisher says training and test rows came from different source families, but chose the final checkpoint by repeated development-set comparison against Jev; this is not a blind general benchmark. The release also trails Jev on several open reasoning benchmarks, including BBH and WinoGrande. The model was trained and evaluated for typed noul and choice questions, returns probabilities instead of chat text, and treats score questions as untested. Its 8,192-token limit applies to each question plus state. The selected 16.55 GB language file excludes the separate 927,607,328-byte F16 vision projector, runtime, context/KV memory and execution headroom; file size alone does not establish speed or compatibility on a computer. The publisher's roughly 55 GB GPU-memory guidance applies to its source-model package, not this Q4 file. Apache-2.0 license.
Context and version evidence: The repository and file name identify the selected model. Revision hashes are shown above when recorded. Tested context length, full hardware configuration and run date are not consistently recorded in this snapshot. Only settings explicitly stated in each benchmark record are known.
The 16.547 GB figure covers model weights in this file. It excludes the context cache, runtime, operating system and any separate multimodal files. It is not a measured RAM requirement or a token-generation speed.
These complete configurations pass the initial weight-size check for the selected 16.55 GB file. Allow extra memory for context and software. Check the model’s runtime notes and your inference software before buying.
Listing evidence: 2026-09-25. Amazon.com base price visible in an owner-supplied screenshot reviewed 25 Sep 2026; capture time not supplied. Dated reference, not a live quote. Excludes card signup offers, coupons, tax and shipping.
Listing evidence: 2026-10-03. Apple US launch price announced 22 September 2026: USD 1,699. Dated manufacturer reference, not an Amazon offer; tax and shipping may differ.
Listing evidence: 2026-09-24. GMKtec, EVO-X2 / 64GB / 1TB, recorded 22 Sep 2026; exact ASIN capacity identified 24 Sep. This is a dated reference, not a current Amazon price; tax, shipping and bundle contents may differ.
Memory matches are estimates, not tested compatibility or speed. As an Amazon Associate I earn from qualifying purchases.
Confirm model architecture and quantization support in your runner. The selected weights use 16.547 GB; this excludes context cache, runtime buffers and other programs.
These are the first listed tiers that pass our reserve policy, not measured minimum requirements. System RAM and VRAM remain separate. Use the memory tool to change the reserve for your context and runner.