Eagle2.5 8B
Apr 2025
Eagle2.5 8BCompare inonMMMU validation
—per task?
self-host onlyprice
—output speed?
—context
Alternatives?
ModelScoreCostPer taskWhat changes
BetterGPT-5.1unstated+53%—$1.25 / $10
OpenMiMo VL 7B RL+20%——Memory (4-bit): 4.8 GB → 4.9 GBBettero3unstated+49%—$2 / $8
BetterClaude Opus 4.5unstated+45%—$5 / $25dBetterDots.vlm1.instOpen+44%——Memory (4-bit): not verifiedModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
GPT-5.1unstated+53%—$1.25 / $10o3unstated+49%—$2 / $8Claude Opus 4.5unstated+45%—$5 / $25dDots.vlm1.instOpen+44%——Memory (4-bit): not verified
Gemini 2.5 Pro Preview (May' 25)unstated+43%——No image inputModelScoreCostPer taskWhat changes
MiMo VL 7B RLOpen+20%——Memory (4-bit): 4.8 GB → 4.9 GBMiMo VL 7B SFTOpen+16%——Memory (4-bit): 4.8 GB → 4.9 GBOInternVL3 8BOpen+12%——Gemma 3 12BunstatedOpen+7%—$0.05 / $0.10Memory (4-bit): 4.8 GB → 6.1 GBOMiniCPM V 2 6Open-11%——Memory (4-bit): 4.8 GB → 4.9 GBModelScoreCostPer taskWhat changes
Nemotron 3.5 Super VLOpenn/a—$0.19Memory (4-bit): 4.8 GB → 75 GBNemotron 3.5 Bolt 74B-A9BOpenn/a——Memory (4-bit): 4.8 GB → 45 GBNVIDIA Nemotron 3.5 Super VL 120B A12B NVFP4Openn/a—$0.08 / $0.45Memory (4-bit): not verifiedNemotron 3.5 LightningthinkingOpenn/a—$0.079Memory (4-bit): 4.8 GB → 20 GB+1Nemotron 3 UltrathinkingOpenn/a—$0.58Memory (4-bit): 4.8 GB → 336 GB+1Nemotron 3.5 Content SafetyOpenn/a—$0.20 / $0.20Nemotron 3 Nano Omni 30B A3B ReasoningOpenn/a—$0 / $0Memory (4-bit): 4.8 GB → 20 GBNVIDIA Nemotron 3 Nano 4BOpenn/a——No image inputAll NVIDIA models →Run it yourself
estimateFits on✓ 8 GB (phone, laptop) · compressed✓ 16 GB · 8-bit✓ 24 GB GPU · full quality✓ 48 GB GPU · full quality✓ 80 GB GPU · full quality
Licencensclv1commercial use with conditions
Benchmarks
in Vision & DocumentsMMMU · validation55.8#27
FAQ
How much does Eagle2.5 8B cost?›
There is no per-unit API price: self-host only.
Is Eagle2.5 8B free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is Eagle2.5 8B, and what is it best at?›
#27 of 41 in Vision & Documents.
Is Eagle2.5 8B better than Qianfan VL 70B?›
On MMMU · validation, Eagle2.5 8B scores 55.8 vs 58.3 — lower.
When was Eagle2.5 8B released? Is there a newer version?›
Released 2025-04-12. It is the current version in its line.
Is Eagle2.5 8B open source?›
Yes, the weights are open (nsclv1 licence).
Can I run Eagle2.5 8B locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Eagle2.5 8B read images?›
It accepts text and images as input.
What is Eagle2.5 8B?›
Eagle2.5 8B is an open-weights language model from NVIDIA, released Apr 2025.
What are the best alternatives to Eagle2.5 8B?›
GPT-5.1 (better), MiMo VL 7B RL (open weights).