Not enough public evidence yet. The creator announced Ajax on . Tom's Hardware reports that it is based on Qwen3.5-9B, but this review could not verify a public Ajax model card, downloadable weight file, license or reproducible benchmark protocol. Do not buy hardware on the assumption that the base model's file size or performance is an Ajax requirement.
Status checked 2026-10-04. Missing evidence is recorded as unavailable, not estimated.
Question
Verified answer
Buying implication
Can I download Ajax?
No creator-linked public weight file or model card was verified in this bounded review.
Wait for an exact repository and revision. Do not download an unrelated file with a similar name.
How much memory does Ajax need?
Unknown. No exact Ajax file, context setting, runtime or peak-memory measurement was verified.
RAM or VRAM capacity alone cannot establish compatibility or speed.
Can I use this Qwen page as a proxy?
Only for the published base model. This page's selected Q4_K_M GGUF is 5.68 GB / 5.29 GiB; it is not an Ajax download or benchmark.
Use it to learn base-model planning, then recheck Ajax after its exact weights and runner instructions are public.
What to verify before choosing a computer
The exact Ajax repository, revision, quantization and complete weight set.
The supported runner and backend, context length, tool interface and any separate multimodal files.
Measured peak memory and tokens per second on the same configuration you intend to buy.
The official Odysseus repository supports local and API models, but its checked README does not supply Ajax weights. Follow the creator announcement for the release context. The October 2 coverage is the source for the reported Qwen3.5-9B relationship; that relationship remains unconfirmed by a public Ajax model card.
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.
MMLU-Pro
82.5 / 100
Knowledge and reasoning across many subjects. Published accuracy; different evaluation settings can change the result.
Benchmark efficiency: 14.523 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-reported results for this exact source model. Evaluation settings differ across model cards; the selected GGUF has not been independently retested. Table row: MMLU-Pro.
Published point estimate, not an independent head-to-head evaluation. Quantization and inference settings can change results.
Hard science questions. This is a science reasoning test, not an overall intelligence score.
Benchmark efficiency: 14.382 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-reported results for this exact source model. Evaluation settings differ across model cards; the selected GGUF has not been independently retested. Table row: GPQA Diamond.
Published point estimate, not an independent head-to-head evaluation. Quantization and inference settings can change results.
Solving coding problems. Version 6 only; published pass rates can vary with generation settings.
Benchmark efficiency: 11.548 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-reported results for this exact source model. Evaluation settings differ across model cards; the selected GGUF has not been independently retested. Table row: LiveCodeBench v6.
Published point estimate, not an independent head-to-head evaluation. Quantization and inference settings can change results.
Following precise instructions. These cards report IFEval without consistently specifying prompt-level versus instruction-level scoring. Treat this view as approximate.
Benchmark efficiency: 16.108 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-reported results for this exact source model. Evaluation settings differ across model cards; the selected GGUF has not been independently retested. Table row: IFEval.
Published point estimate, not an independent head-to-head evaluation. Quantization and inference settings can change results.
Choosing and calling functions. Published BFCL v4 scores; tool environments and settings differ.
Benchmark efficiency: 11.636 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-reported results for this exact source model. Evaluation settings differ across model cards; the selected GGUF has not been independently retested. Table row: BFCL-V4.
Published point estimate, not an independent head-to-head evaluation. Quantization and inference settings can change results.
Revised MMLU knowledge test. Kept separate from original MMLU and MMLU-Pro.
Benchmark efficiency: 16.037 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-reported results for this exact source model. Evaluation settings differ across model cards; the selected GGUF has not been independently retested. Table row: MMLU-Redux.
Published point estimate, not an independent head-to-head evaluation. Quantization and inference settings can change results.
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 5.681 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 5.68 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 5.681 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.