Skip to content
Model Pareto

Falcon-H1R-7B

TII (Falcon)Open weightsartificialanalysis.ai ↗
Input
—
Output
—
Speed
~99tok/sest.
Context
256Ktokens

Long-tail import from Artificial Analysis (2026-09-24); AA variant: Falcon-H1R-7B. Price = AA median host price.

Rankings

Self-hostingEstimated

How we estimate →

What it would cost to run these open weights yourself on rented GPUs. No API sells this model, so this is its price on the chart.

Hardware
1× B200 180GB
BF16 weights
Throughput
~11,082 tok/s
many requests batched
Price
$0.074–0.66 /M tok
busy → light use
On your own machine
RTX 5090 32GB (int4, 300+ tok/s single-stream)
single consumer GPU or Mac
Assumptions (5)
  • BF16 weights (7.59B params) + 25% KV-cache headroom ≈ 19 GB
  • 1× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
  • ~11,082 output tok/s aggregate at batch 256 (bandwidth-bound)
  • Blended 3:1 input:output; prefill ~70,035 tok/s
  • Low = 75% utilization at the low GPU price; high = 20% at the high price

Benchmark scores

~ italic, dashed = no published score yet, estimated from related benchmarks.

Writing & chat

MMMLUEstimated
~last of 5 (estimated)below measured range (<81.3%)
  1. #1Gemini 3.1 Pro92.6%
  2. #2Gemma 4 31B88.4%
  3. #4gpt-oss-120b81.3%
  4. ~#5Falcon-H1R-7BEstimated<81.3%
No published score — estimated from related benchmarks
~#45 of 57 (estimated)~1376 Elo
  1. #1Gemini 4 Argon1525
  2. #2Claude Opus 5.51507
  3. #44GLM-4.6VFew votes1378
  4. ~#45Falcon-H1R-7BEstimated1376
  5. #45Trinity Large Thinking1369
No published score — estimated from related benchmarks
~#19 of 26 (estimated)~1560 Elo
  1. #1GPT-6 Astra2173
  2. #2Claude Fable 5.12162
  3. #18Inkling1611
  4. ~#19Falcon-H1R-7BEstimated1560
  5. #19Gemini 3.5 Flash-Lite1559
No published score — estimated from related benchmarks

Hard reasoning

Agents

~#12 of 15 (estimated)~55.3%
  1. #1Kimi K384.8%
  2. #2Qwen3.8 27B84.3%
  3. #11Muse Glimmer65.9%
  4. ~#12Falcon-H1R-7BEstimated55.3%
  5. #12Nemotron 3.5 Super VL NVFP452.3%
No published score — estimated from related benchmarks
OSWorld 2.0Estimated
~#3 of 5 (estimated)~53.8%
  1. #1GLM-5.3 Flash59.1%
  2. #2Kimi K358.3%
  3. ~#3Falcon-H1R-7BEstimated53.8%
  4. #3GPT-5.6 Terra50.2%
No published score — estimated from related benchmarks
BrowseCompEstimated
~#21 of 22 (estimated)~41.4%
  1. #1Atria Dawn Preview92.5%
  2. #2GPT-6 Astra91.5%
  3. #20Nemotron 3 Ultra44.4%
  4. ~#21Falcon-H1R-7BEstimated41.4%
  5. #21Mistral Small 421.3%
No published score — estimated from related benchmarks
~#55 of 92 (estimated)~757 Elo
  1. #1Claude Opus 5.51866
  2. #2Muse Wonder Full1857
  3. #54Gemini 3.1 Pro776
  4. ~#55Falcon-H1R-7BEstimated757
  5. #55Qwen3.7 Plus756
No published score — estimated from related benchmarks
~#50 of 85 (estimated)~6.1%
  1. #1Claude Fable 5.276.4%
  2. #2Muse Wonder Full71.3%
  3. #49Mistral Medium 3.56.3%
  4. ~#50Falcon-H1R-7BEstimated6.1%
  5. #50GPT-5.4 nano6.0%
No published score — estimated from related benchmarks
~#63 of 91 (estimated)~10.4%
  1. #1Muse Spark 1.350.5%
  2. #2GLM-5.350.3%
  3. #62Devstral 210.5%
  4. ~#63Falcon-H1R-7BEstimated10.4%
  5. #63Nemotron 3 Super 120B A12B10.3%
No published score — estimated from related benchmarks