Model catalog & specs
Waired bundles a catalog of models spanning laptops to multi-GPU servers. On first run it auto-selects one that fits your hardware. You can switch at any time — see Switch the bundled model.
Finding the right table
Section titled “Finding the right table”The catalog is split two ways so you can jump straight to the rows that apply to your machine:
-
By runtime — Ollama vs vLLM. Your OS and GPU decide which engine serves the model, so each runtime has its own section. Read only yours:
- Ollama — macOS (Metal), Windows, Linux, CPU, and low-VRAM / integrated GPUs. Ships GGUF (quantized) builds; the number to watch is system RAM.
- vLLM — NVIDIA (CUDA) and AMD (ROCm) discrete-GPU servers on Linux. Ships safetensors (awq/fp8/mxfp4) builds; the number to watch is GPU VRAM.
A family that ships builds for both engines appears in both sections.
-
By architecture — Dense vs MoE. Within each runtime, Dense models come first, then Mixture-of-Experts (MoE):
- Dense — every parameter runs on every token. Best where you have compute / VRAM headroom (a discrete GPU), and smaller on disk for a given quality.
- MoE — huge total size but only a few billion parameters are active per token, so decode stays fast. Best on memory-rich unified-memory machines (Apple Silicon, AMD Strix Halo) that can hold a large model in RAM.
Quality tier is a 1–100 ranking (higher is better); the Auto-Selector serves
the highest tier that fits your machine and can hold a large coding context
(~200k tokens) — coding agents routinely fill 100k+ tokens of context, so a
model that would truncate it loses to a slightly smaller one with the full
window. On hardware where no model can serve that window, Waired picks the
best fit anyway and says so in waired status. Whatever wins, clients can
address it by the stable handle waired/default (see
Chat clients).
The tables below are generated at build time from each model’s bundled manifest
(min_ram_gb / min_vram_mb / quality_tier / param_count / active_params),
so they always match the catalog the client ships.
Ollama models
Section titled “Ollama models”For Mac, Windows, CPU, and low-VRAM / integrated GPUs. The memory column is the minimum system RAM for the GGUF (q4) build; smaller machines are listed first.
Ollama · Dense
Section titled “Ollama · Dense”Every parameter runs each token — pick by RAM, then quality.
| Model | Params | Quality tier | Ollama RAM |
|---|---|---|---|
qwen2.5-coder-0.5b-instruct | 494M | 10 | 2 GB |
qwen3.5-0.8b | 800M | 12 | 2 GB |
qwen2.5-coder-3b-instruct | 3.1B | 31 | 4 GB |
qwen3.5-2b | 2B | 27 | 4 GB |
qwen2.5-coder-7b-instruct | 7.6B | 50 | 8 GB |
qwen3.5-4b | 4B | 42 | 8 GB |
qwen3.5-9b | 9B | 52 | 12 GB |
qwen2.5-coder-14b-instruct | 14.7B | 58 | 16 GB |
qwen3.5-27b | 27B | 69 | 24 GB |
qwen3.6-27b | 27B | 72 | 24 GB |
Ollama · MoE
Section titled “Ollama · MoE”Large on disk but fast to decode — ideal for big unified-memory machines (Apple Silicon, Strix Halo).
| Model | Params (active) | Quality tier | Ollama RAM |
|---|---|---|---|
gpt-oss-20b | 20.9B (3.6B active) | 62 | 16 GB |
qwen3-coder-30b-a3b-instruct | 30.5B (3.3B active) | 68 | 32 GB |
qwen3.5-35b-a3b | 35B (3.3B active) | 73 | 32 GB |
qwen3.6-35b-a3b | 35B (3.3B active) | 90 | 32 GB |
qwen3-coder-next-80b-a3b-instruct | 80.1B (3.3B active) | 86 | 56 GB |
gpt-oss-120b | 117B (5.1B active) | 88 | 96 GB |
qwen3.5-122b-a10b | 122B (10B active) | 83 | 128 GB |
qwen3-coder-480b-a35b-instruct | 480B (35B active) | 95 | 320 GB |
vLLM models
Section titled “vLLM models”For NVIDIA (CUDA) and AMD (ROCm) discrete-GPU servers on Linux. The memory column
is the minimum GPU VRAM for the quantized (awq/fp8/mxfp4) build; smaller GPUs
are listed first. On hosts with several identical NVIDIA GPUs, vLLM shards the
model across them (tensor parallelism), so their combined VRAM counts toward
the requirement. On recent NVIDIA GPUs (Ada/Hopper and newer — e.g. L4, RTX
40-series), Waired serves the KV cache in fp8, which roughly doubles the context
window that fits in the same VRAM at near-identical quality. If a model’s full
context window still doesn’t fit your GPU memory, Waired serves it with a reduced
window and notes that in waired status / waired doctor instead of failing to
start.
vLLM · Dense
Section titled “vLLM · Dense”Compute-bound; the natural fit when you have a discrete GPU with VRAM headroom.
| Model | Params | Quality tier | vLLM VRAM |
|---|---|---|---|
qwen2.5-coder-3b-instruct | 3.1B | 31 | 4 GB |
qwen2.5-coder-7b-instruct | 7.6B | 50 | 8 GB |
qwen2.5-coder-14b-instruct | 14.7B | 58 | 16 GB |
qwen3.6-27b | 27B | 72 | 24 GB |
vLLM · MoE
Section titled “vLLM · MoE”Big total size with a small active footprint — pairs well with multi-GPU servers.
| Model | Params (active) | Quality tier | vLLM VRAM |
|---|---|---|---|
gpt-oss-20b | 20.9B (3.6B active) | 62 | 20 GB |
qwen3-coder-30b-a3b-instruct | 30.5B (3.3B active) | 68 | 24 GB |
qwen3-coder-next-80b-a3b-instruct | 80.1B (3.3B active) | 86 | 55 GB |
gpt-oss-120b | 117B (5.1B active) | 88 | 79 GB |
glm-4.5-air-106b-a12b | 106B (12B active) | 75 | 118 GB |
deepseek-v4-flash | 284B (13B active) | 93 | 192 GB |
glm-5.2 | 744B (40B active) | 97 | 547 GB |
qwen3-coder-480b-a35b-instruct | 480B (35B active) | 95 | 547 GB |
How Waired picks for you
Section titled “How Waired picks for you”The Auto-Selector chooses the highest-quality model that fits your machine’s memory. On shared-memory systems (Apple Silicon, AMD Strix Halo), MoE models with a small active parameter count are favored because decode speed tracks active params, not total size. To preview a routing decision without changing anything:
waired infer --explain "say hi"Run waired models ls --detail to see this same fit + recommended-spec view for
your hardware, waired models ls for disk usage, and waired runtimes status
for the VRAM a loaded model is actually using.
If you run an integrated GPU (Strix Halo, Intel) and a model won’t load on Ollama, see Troubleshooting → integrated GPUs.