Compare · ModelsLive · 2 picked · head to head

Phi-1.5 vs Cerebras-GPT-13B

Side by side · benchmarks, pricing, and signals you can act on.

Winner summary

Phi-1.5 wins 4 of 5 shared benchmarks. Leads in knowledge.

Category leads
knowledge·Phi-1.5
Hype vs Reality
Phi-1.5
#221 by perf·no signal
QUIET
Cerebras-GPT-13B
#209 by perf·no signal
QUIET
Best value
Phi-1.5
no price
Cerebras-GPT-13B
no price
Vendor risk
Microsoft logo
Microsoft
$3.00T·Big Tech
Low risk
OpenAI logo
OpenAI
$840.0B·Tier 1
Medium risk
Head to head
Phi-1.5Cerebras-GPT-13B
ARC AI2
Phi-1.5 leads by +16.0
AI2 Reasoning Challenge · tests grade-school level science knowledge with multiple-choice questions requiring reasoning beyond simple retrieval.
Phi-1.5
25.9
Cerebras-GPT-13B
9.9
HellaSwag
Cerebras-GPT-13B leads by +15.7
HellaSwag · tests commonsense reasoning by asking models to predict the most plausible continuation of everyday scenarios.
Phi-1.5
30.1
Cerebras-GPT-13B
45.9
MMLU
Phi-1.5 leads by +15.2
Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge.
Phi-1.5
16.8
Cerebras-GPT-13B
1.6
OpenBookQA
Phi-1.5 leads by +1.9
OpenBookQA · science questions that require combining a given core fact with broad common knowledge, mimicking an open-book exam setting.
Phi-1.5
16.3
Cerebras-GPT-13B
14.4
Winogrande
Phi-1.5 leads by +25.2
WinoGrande · large-scale commonsense reasoning benchmark where models must resolve ambiguous pronouns in carefully constructed sentence pairs.
Phi-1.5
46.8
Cerebras-GPT-13B
21.6
Full benchmark table
BenchmarkPhi-1.5Cerebras-GPT-13B
ARC AI2
AI2 Reasoning Challenge · tests grade-school level science knowledge with multiple-choice questions requiring reasoning beyond simple retrieval.
25.99.9
HellaSwag
HellaSwag · tests commonsense reasoning by asking models to predict the most plausible continuation of everyday scenarios.
30.145.9
MMLU
Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge.
16.81.6
OpenBookQA
OpenBookQA · science questions that require combining a given core fact with broad common knowledge, mimicking an open-book exam setting.
16.314.4
Winogrande
WinoGrande · large-scale commonsense reasoning benchmark where models must resolve ambiguous pronouns in carefully constructed sentence pairs.
46.821.6
Pricing · per 1M tokens · projected $/mo at 10M tokens
ModelInputOutputContextProjected $/mo
Microsoft logoPhi-1.5
OpenAI logoCerebras-GPT-13B