llmrun
InternLMQwen3_5ForConditionalGeneration

Intern Decision 0.8B — Hardware Requirements & GPU Compatibility

VisionFunctions

Intern-Decision-0.8B comes from InternLM as a 0.8-billion-parameter multimodal structured decision model, fine-tuned from Qwen3.5-0.8B. Instead of generating free text, it takes a shared state, a schema of named questions and optional images, then returns an answer distribution for every question in one forward pass by reading logits over candidate option symbols. This suits classification and typed-decision tasks; the card notes scores of 94.52 on ToolACE and 77.35 on Typed Decision. It handles up to eight images per request, and at this size it runs on any consumer GPU or CPU. Context length reaches 262,144 tokens, with image tokens counted toward the input limit. Released under the Apache 2.0 license for unrestricted commercial and research use, it appeared in September 2026. Built on a Qwen base as a specialized scorer, this model is not intended for general chat.

853M params · dense262K contextApache 2.03.7K downloads38 likes792 quant downloads
Based on Qwen3.5 0.8B
S

Runs great on a 24 GB GPU

Runs from8 GB Q4_K_M · 256KBest on24 GB Q4_K_M · ~762 tok/s · 256KMacs from8 GB Q4_K_M · ~83 tok/s

Apache 2.0

Reference GPU · RTX 4090 24 GB

Estimates, ±20% · how we estimate

Run Intern Decision 0.8B

One recommended way. Every other method, OS and quant is in the step-by-step guide.Each way opens a step-by-step guide for your OS and quant.

Recommended

Ollama
one command · needs Ollama installed
$ollama run hf.co/bombdefuser-124/Intern-Decision-0.8B-GGUF

Q4_K_M fits fully on the RTX 4090 · 23.1 GB free · 16K context

Download: Intern-Decision-0.8B-Q4_K_M.gguf0.51 GBbombdefuser-124

install, OS, other quants

Ways to runOther ways

each opens step-by-step

Run in cloud

Runs on RTX 3060 12GB (11 GB headroom) · Q4_K_M

Generation speed
~272 tok/s
generation speed
Cost per 1M output tokens
$0.064
per 1M output tokens

How Much VRAM Does Intern Decision 0.8B Need?

Select a quantization level; the listed hardware adheres to it.

QuantizationVRAM+32KOn RTX 4090 (reference)
0.7 GB0.9 GB2-bit · K-quant — 2-bit quantization with K-quant improvementsFits: 256K context · ~923 tok/s
0.8 GB0.9 GB3-bit medium — 3-bit medium quantizationFits: 256K context · ~851 tok/s
0.9 GB1 GB4-bit medium — 4-bit medium quantization — most popular sweet spotFits: 256K context · ~762 tok/s
1.0 GB1.1 GB5-bit medium — 5-bit medium quantization — good quality/size tradeoffFits: 256K context · ~683 tok/s
1.1 GB1.2 GB6-bit, very good — 6-bit quantization, very good qualityFits: 256K context · ~624 tok/s
1.2 GB1.4 GB8-bit, near-lossless — 8-bit quantization, near-losslessFits: 256K context · ~546 tok/s
2.1 GB2.2 GB16-bit · bfloat — Brain floating point 16 — preferred for trainingFits: 256K context · ~318 tok/s
est.calculated, no published file yet

Selected Q4_K_M

4-bit medium quantization

0.9GBat 16K context

Weights
0.5 GB
KV cache
0.1 GB
Overhead
0.3 GB

Fits RTX 4090 (reference) · 23.1 GB free · ~762 tok/s

≥50 tok/s needs ≥66 GB/s · RTX 5090 ~1354, M5 Ultra ~1000

Which Hardware Can Run Intern Decision 0.8B?

Q4_K_M · 0.9 GB

Intern Decision 0.8B (Q4_K_M) requires 0.9 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 2+ GB is recommended. Using the full 262K context window can add up to 1.6 GB, bringing total usage to 2.4 GB. 114 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 185 devices with unified memory can run Intern Decision 0.8B, including Mac Studio M5 Ultra (256 GB), Mac Studio M5 Ultra (512 GB).

By VRAM tier · typical GPUs and Macs at Q4_K_M
8 GB · RTX 4060Q4_K_M · 0.9 GB · ~206 tok/s · 256K8 GB tier →
12 GB · RTX 3060 12GBQ4_K_M · 0.9 GB · ~272 tok/s · 256K12 GB tier →
16 GB · RTX 5060 Ti 16GBQ4_K_M · 0.9 GB · ~339 tok/s · 256K16 GB tier →
24 GB · RTX 4090referenceQ4_K_M · 0.9 GB · ~762 tok/s · 256K24 GB tier →
Mac 32 GB · Mac Mini M4Q4_K_M · 0.9 GB · ~98 tok/s · 256KMacs →
Mac 64 GB · Mac Studio M4 MaxQ4_K_M · 0.9 GB · ~444 tok/s · 256KMacs →

Alternatives to Intern Decision 0.8B

Similar models

1
ModelScoreVRAMtok/sWhy it is similar
Qwen3.5 0.8BAlibaba—0.9 GB~753same size

Specifications

Publisher
InternLM
Parameters
853M
Architecture
Qwen3_5ForConditionalGeneration
Context Length
262,144 tokens
Vocabulary Size
248,320
Release Date
2026-09-26
License
Apache 2.0

Frequently Asked Questions

How much VRAM does Intern Decision 0.8B need?

Intern Decision 0.8B requires 0.9 GB of VRAM at Q4_K_M, or 2.1 GB at BF16. Full 262K context adds up to 1.6 GB (2.4 GB total).

VRAM = Weights + KV Cache + Overhead

Weights = 853M × 4.8 bits ÷ 8 = 0.5 GB

KV Cache + Overhead ≈ 0.4 GB (at 2K context + ~0.3 GB framework)

Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.

KV Cache + Overhead ≈ 1.9 GB (at full 262K context)

VRAM usage by quantization

0.9 GB
2.4 GB

Learn more about VRAM estimation →

What's the best quantization for Intern Decision 0.8B?

For Intern Decision 0.8B, Q4_K_M (0.9 GB) offers the best balance of quality and VRAM usage. Q5_K_M (1.0 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 0.7 GB.

VRAM requirement by quantization

Q2_K
0.7 GB
Q4_K_M ★
0.9 GB
Q5_K_M
1.0 GB
Q6_K
1.1 GB
Q8_0
1.2 GB
BF16
2.1 GB

★ Recommended — best balance of quality and VRAM usage.

Learn more about quantization →

Can I run Intern Decision 0.8B on a Mac?

Yes — MacBook Air 13" M3 (8 GB) and 158 other Macs can run Intern Decision 0.8B. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 0.9 GB of usable unified memory (RAM minus macOS overhead).

Can I run Intern Decision 0.8B locally?

Yes — Intern Decision 0.8B can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.

How fast is Intern Decision 0.8B?

At Q4_K_M, Intern Decision 0.8B can reach ~5581 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~762 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.

tok/s = (bandwidth GB/s ÷ model GB) × efficiency

Example: NVIDIA B200 → 8000 ÷ 0.86 × 0.65 = ~6047 tok/s

Estimated speed at Q4_K_M (0.9 GB)

~6047 tok/s
~762 tok/s
~6047 tok/s
~5581 tok/s

Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.

Learn more about tok/s estimation →

What's the download size of Intern Decision 0.8B?

At Q4_K_M, the download is about 0.51 GB. The full-precision BF16 version is 1.71 GB. The smallest option (Q2_K) is 0.36 GB.

Which GPUs can run Intern Decision 0.8B?

114 consumer GPUs can run Intern Decision 0.8B at Q4_K_M. Top options include AMD Radeon RX 5600 XT, AMD Radeon RX 5700 XT, AMD Radeon RX 580 8GB. 114 GPUs have plenty of headroom for comfortable inference.

Which devices can run Intern Decision 0.8B?

185 devices with unified memory can run Intern Decision 0.8B at Q4_K_M (0.9 GB), including AMD Ryzen 7 8845HS (Hawk Point) Laptop, AMD Ryzen AI 7 350 (Krackan Point) Laptop, AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10. Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.