Falcon H1R 7B

Name: Falcon H1R 7B
Author: TII (Falcon)

TII (Falcon) · Falcon-H1 · released 2026-01-05 · data checked 2026-11-01 Needs verification

Falcon H1R 7B is a 7.6B-parameter dense model from TII (Falcon) (Other), released 2026-01-05 under the Falcon LLM License license. It scores 57.5 on the Small LM Index, #5 of 26 small models. It needs about 5.6 GB of memory at 4-bit with an 8K-token context.

Hybrid Transformer/Mamba reasoning model. Release date approximate (weights uploaded Oct 2025, announced Jan 2026).

Small LM Index

57.5

#5 of 26 scored small models

Likely 54.2–60.7 · from 4 of 6 core benchmarks

Parameters

7.6B

Dense

Price per 1M tokens

—

No hosted API on OpenRouter; self-host

Context window

262K

tokens

Benchmarks

Each bar shows this model's score. The tick marks the best small model on that benchmark. Sources: 7 reported by TII (Falcon). How we label sources.

MMLU-Pro Lab-reported Knowledge72.1

Class best: Qwen3.5 9B 82.5

GPQA Diamond Lab-reported Knowledge61.3

Class best: Qwen3.5 9B 81.7

Humanity's Last Exam Lab-reported Reasoning11.1

AIME Lab-reported Math · AIME 202583.1

Class best: Apriel 1.6 15B Thinker 88.0

IFBench Lab-reported Instruction following53.4

Class best: Granite 4.2 8B 79.3

MATH-500 Lab-reported Math97.4

HMMT Lab-reported Math · HMMT Feb 202564.9

Class best: Granite 4.2 8B 78.3

Not reported: estimated

What the index model expects from this model's other scores. Not a measurement.

LiveCodeBench≈ 66

SWE-bench Verified≈ 42

Specifications

Lab: TII (Falcon) (AE · Other)
Architecture: Dense transformer
Total parameters: 7.6B
Active parameters: All (dense)
Size class: Small
Context window: 262K tokens
Input modalities: text
Reasoning: Reasoning (thinking)
License: Falcon LLM License
Openness: Open weights
Memory to run: ~5.6 GB at 4-bit, 8K context (more sizes)
Hugging Face: 2.4K downloads (30 days) · 222 likes

Weights on Hugging Face ↗

Something wrong here? Report it.

Run it locally

Uses unsloth/Falcon-H1R-7B-GGUF (Q4_K_M). Swap the tag for another size: BF16, IQ4_NL, IQ4_XS, Q2_K, Q2_K_L, Q3_K_M, Q3_K_S, Q4_0….

Ollama

ollama run hf.co/unsloth/Falcon-H1R-7B-GGUF:Q4_K_M

llama.cpp

llama-server -hf unsloth/Falcon-H1R-7B-GGUF:Q4_K_M

Then open http://localhost:8080 for a chat window.

MLX (Mac)

pip install -U mlx-lm
mlx_lm.chat --model mlx-community/Falcon-H1R-7B-4bit

vLLM (GPU server)

vllm serve tiiuae/Falcon-H1R-7B

Full-precision weights, for a server GPU.

Open in LM Studio

Memory needed: weights, context cache and runtime buffers.

Quantization	4K context	32K context	256K context
16-bit F16	16 GB	17 GB	28 GB
8-bit Q8_0	8.9 GB	10 GB	21 GB
4-bit Q4_K_M	5.4 GB	6.7 GB	17 GB

Check it against your GPU, Mac or phone →

Falcon H1R 7B

Benchmarks

Not reported: estimated

Specifications

Compare with

Ranked in

Sources

Run it locally