Models / Pokee-Isaac 28B

Pokee-Isaac 28B

enthusiast

28B-parameter proprietary non-decoder-only architecture from Pokee AI. The model is pitched as the first real 10M-token context frontier-class agentic model that fits on a single GPU - starting from an RTX 4090 (24GB) for the quantized form.

Vendor-reported highlights. 93.3% RULER at 10M-token context; up to 137K tokens/s prefill on one B200 with a 10M-token context; leads BFCL v4 and tau^3-bench in Pokee’s evaluation; lowest combined attack success rate among evaluated models on the DTAP security red-teaming benchmark.

Deployment and pricing. Closed weights. Hosted API is $0.15/M input, $1/M output. Also deployable in your VPC, on-premises, or on-device with Day-0 support for vLLM and SGLang. Technical docs and API console live at https://console.pokee.ai.

Honest framing. The architecture and most benchmarks are vendor-reported; the 10M context and single-GPU deployment claim are the concrete differentiators to watch as independent evals appear. The 28B size is consistent with the stated RTX 4090 footprint, but treat the throughput and safety numbers as claims until replicated.

coding reasoning agentic
Parameters
28.0B
Context
10000k
License
other
Developer
Pokee AI
Origin
🇺🇸 USA
Released
Aug 2026

What people are building with Pokee-Isaac 28B

Real demos from X

Launch post: 28B, 10M context, single-GPU deployment, $0.15/$1 per 1M tokens View on X →

Scores

Coding
90
Reasoning
88
Tool calling
95
General
88

Score per dollar

587 pts per $/M input

general_score (88) divided by cheapest input price ($0.15/M). Higher is better value. See live pricing.

Guides covering Pokee-Isaac 28B

Save your hardware and every model page answers the real question: will it run on your machine, and how fast?

Join free - save your rig →

Run it locally

Per-quant memory needs and a static "can you run it?" reference - no rig entry required

Q4_K_M
16.0GB 18.0GB min 22.0GB rec
Balanced - the usual local sweet spot
Q8_0
28.0GB 30.0GB min 32.0GB rec
Near-lossless
FP16
56.0GB 56.0GB min 64.0GB rec
Full quality, largest

The reference hardware

Schematic of the NVIDIA Jetson Orin NX 16GB reference rig - 16GB unified memory, 102 GB/s aggregate bandwidth
Schematic of the Single GTX 1080 Ti (11GB) reference rig - 11GB VRAM, 484 GB/s aggregate bandwidth
Schematic of the 4x H100 80GB (320GB) reference rig - 320GB VRAM, 13400 GB/s aggregate bandwidth
Schematic of the NVIDIA DGX Station 748GB reference rig - 748GB unified memory, 8000 GB/s aggregate bandwidth
Schematic of the 8x RTX 3090 rack (192GB) reference rig - 192GB VRAM, 7489 GB/s aggregate bandwidth
Schematic of the 4x RTX 5090 (128GB) reference rig - 128GB VRAM, 7168 GB/s aggregate bandwidth
Schematic of the AMD Instinct MI300X (192GB) reference rig - 192GB VRAM, 5324 GB/s aggregate bandwidth
Schematic of the 4x RTX 4090 (96GB) reference rig - 96GB VRAM, 4032 GB/s aggregate bandwidth
Schematic of the 2x RTX 5090 (64GB) reference rig - 64GB VRAM, 3584 GB/s aggregate bandwidth
Schematic of the 2x RTX 3090 (48GB) reference rig - 48GB VRAM, 1872 GB/s aggregate bandwidth
Schematic of the Single RTX 5090 (32GB) reference rig - 32GB VRAM, 1792 GB/s aggregate bandwidth
Schematic of the RTX PRO 6000 Blackwell (96GB) reference rig - 96GB VRAM, 1792 GB/s aggregate bandwidth
Schematic of the Mac Studio M4 Ultra 192GB reference rig - 192GB unified memory, 1092 GB/s aggregate bandwidth
Schematic of the Mac Studio M4 Ultra 512GB reference rig - 512GB unified memory, 1092 GB/s aggregate bandwidth
Schematic of the Single RTX 4090 (24GB) reference rig - 24GB VRAM, 1008 GB/s aggregate bandwidth
Schematic of the MacBook Pro M5 Max 128GB reference rig - 128GB unified memory, 614 GB/s aggregate bandwidth
Schematic of the Dual EPYC 9004 + 768GB DDR5-4800 reference rig - 768GB unified memory, 460 GB/s aggregate bandwidth
Schematic of the DGX Spark 128GB unified reference rig - 128GB unified memory, 273 GB/s aggregate bandwidth
Schematic of the Ryzen AI Max+ 395 128GB reference rig - 128GB unified memory, 256 GB/s aggregate bandwidth
Schematic of the Jetson AGX Orin 64GB reference rig - 64GB unified memory, 204 GB/s aggregate bandwidth
Schematic of the Epyc + 512GB DDR4-3200 + 2x RTX 3090 reference rig - 560GB unified memory, 204 GB/s aggregate bandwidth
Schematic of the Epyc + 512GB DDR4-2400 + 2x RTX 3090 reference rig - 560GB unified memory, 153 GB/s aggregate bandwidth

22 reference configs, drawn in-house. Scroll for more.

Can you run it? - reference rigs

Rig Q4_K_M Q8_0 FP16
NVIDIA Jetson Orin NX 16GB no -> cloud no -> cloud no -> cloud
Single GTX 1080 Ti (11GB) offload no -> cloud no -> cloud
4x H100 80GB (320GB) fast 460.6t/s fast 263.2t/s fast 131.6t/s
NVIDIA DGX Station 748GB fast 275.0t/s fast 157.1t/s fast 78.6t/s
8x RTX 3090 rack (192GB) fast 257.4t/s fast 147.1t/s fast 73.6t/s
4x RTX 5090 (128GB) fast 246.4t/s fast 140.8t/s fast 70.4t/s
AMD Instinct MI300X (192GB) fast 183.0t/s fast 104.6t/s fast 52.3t/s
4x RTX 4090 (96GB) fast 138.6t/s fast 79.2t/s fast 39.6t/s
2x RTX 5090 (64GB) fast 123.2t/s fast 70.4t/s fast 35.2t/s
2x RTX 3090 (48GB) fast 64.4t/s fast 36.8t/s offload
Single RTX 5090 (32GB) fast 61.6t/s fast 35.2t/s no -> cloud
RTX PRO 6000 Blackwell (96GB) fast 61.6t/s fast 35.2t/s ok 17.6t/s
Mac Studio M4 Ultra 192GB fast 40.9t/s fast 23.4t/s ok 11.7t/s
Mac Studio M4 Ultra 512GB fast 40.9t/s fast 23.4t/s ok 11.7t/s
Single RTX 4090 (24GB) fast 34.6t/s offload no -> cloud
MacBook Pro M5 Max 128GB fast 23.0t/s ok 13.2t/s slow 6.6t/s
Dual EPYC 9004 + 768GB DDR5-4800 ok 15.8t/s ok 9.1t/s slow 4.5t/s
DGX Spark 128GB unified ok 9.4t/s slow 5.4t/s slow 2.7t/s
Ryzen AI Max+ 395 128GB ok 8.8t/s slow 5.0t/s slow 2.5t/s
Jetson AGX Orin 64GB slow 7.0t/s slow 4.0t/s slow 2.0t/s
Epyc + 512GB DDR4-3200 + 2x RTX 3090 slow 7.0t/s slow 4.0t/s slow 2.0t/s
Epyc + 512GB DDR4-2400 + 2x RTX 3090 slow 5.3t/s slow 3.0t/s slow 1.5t/s

Fit tiers use the same will-it-run logic as the rig finder. For comfortable fits, the badge reflects decode speed: fast >=20 t/s, ok 8-20 t/s, slow <8 t/s. t/s is a bandwidth estimate, not a measured benchmark.

Download options

Q4_K_M official -5% vs fp16
16.0GB dl 18.0GB min 22.0GB rec
REC RAM vs largest quant
Q8_0 official -1% vs fp16
28.0GB dl 30.0GB min 32.0GB rec
REC RAM vs largest quant
FP16 official
56.0GB dl 56.0GB min 64.0GB rec
REC RAM vs largest quant

Or run it in the cloud

Live per-provider pricing, throughput and uptime - refreshed about 2 months ago via OpenRouter. Click a column to sort.

some pricing may be stale - last verified 2026-08-05

Provider Type Input $/M Output $/M Cache $/M Tok/s Latency Uptime Value
Pokee AI stale
API 0.15 1.00 - - - - cheapest

Default order: throughput among 95%+ uptime providers, then latency; subscriptions last. Sort by any column. Subscription rows show $/mo in the Value column - per-token columns are "-". Affiliate links are marked sponsored / nofollow. Confirm current pricing on the provider's site before committing.

Detailed API pricing page + JSON endpoint →

PRICE HISTORY

Inference cost over time

Data accumulates from the first daily sync - longer ranges populate over time. Prices come from OpenRouter snapshots, not a historical API.

Loading price history...