vs
0:1
overall capability wins (2 ties across 3 scored). Yi Lightning leads overall.
- Arena0:1
- Pricing0:3
- Speed0:2
Verdict
Yi Lightning ranks higher on LMArena (+116 Elo); Yi-1.5 34B Chat ships open weights (Apache 2.0); Yi Lightning is ~3.6x cheaper on a blended token basis than Yi-1.5 34B Chat.
- Yi Lightning ranks higher on LMArena (+116 Elo)
- Yi-1.5 34B Chat ships open weights (Apache 2.0)
- Yi Lightning is ~3.6x cheaper on a blended token basis than Yi-1.5 34B Chat
Cost for 1M in + 250K out
Illustrative chat workload at primary-provider list prices.
- Yi-1.5 34B Chat
- $0.63
- Yi Lightning
- $0.18
Yi Lightning is about 3.6x cheaper on this workload.
Price per 1M tokens
Full comparison
Organization
01.AI
01.AI
Family
Yi
Yi
License
Apache 2.0
Proprietary
Open weights
Yes
No
Release
May 20, 2024
Jan 15, 2025
Knowledge cutoff
-
-
API / provider
01.AI
01.AI
Modalities
text → text
text → text
Specs
Context window
16K
16K
Max output
4K
4K
Parameters
34B
—
Pricing
Input $/1M
$0.50
$0.14
Output $/1M
$0.50
$0.14
Blended $/1M (3∶1)
$0.50
$0.14
Speed
tok/s
55
200
TTFT (s)
0.45
0.15
Reasoning
MMLU-Pro
—
—
GPQA Diamond
—
—
Humanity's Last Exam
—
—
AIME 2025
—
—
MATH-500
—
—
Humanity's Last Exam (with tools)
—
—
AA-Omniscience Accuracy
—
—
AA-LCR v1.1
—
—
CritPt
—
—
MMMU-Pro
—
—
IFBench
—
—
Chartography
—
—
Chartography (With Tools)
—
—
Coding
SWE-bench Verified
—
—
SWE-bench Pro
—
—
SWE-bench Multilingual
—
—
LiveCodeBench
—
—
Terminal-Bench 2.1
—
—
Aider Polyglot
—
12.9
Terminal-Bench 3
—
—
BigCodeBench
33.9
—
SciCode
—
—
CursorBench
—
—
SWE-Rebench
—
—
NL2Repo-Bench
—
—
DeepSWE
—
—
WebDev Arena
—
—
Terminal-Bench 4.0
—
—
CursorBench 4.0
—
—
FrontierCode v1.1 (Main)
—
—
Arena
LMArena Elo
1,213
1,328
| Metric | Yi-1.5 34B Chat | Yi Lightning | Delta |
|---|---|---|---|
| Identity | |||
| Organization | 01.AI | 01.AI | — |
| Family | Yi | Yi | — |
| License | Apache 2.0 | Proprietary | — |
| Open weights | Yes | No | — |
| Release | May 20, 2024 | Jan 15, 2025 | — |
| Knowledge cutoff | - | - | — |
| API / provider | 01.AI | 01.AI | — |
| Modalities | text → text | text → text | — |
| Specs | |||
| Context window | 16K | 16K | tie |
| Max output | 4K | 4K | tie |
| Parameters | 34B | — | — |
| Pricing | |||
| Input $/1M | $0.50 | $0.14 | B +$0.36 |
| Output $/1M | $0.50 | $0.14 | B +$0.36 |
| Blended $/1M (3∶1) | $0.50 | $0.14 | B +$0.36 |
| Speed | |||
| tok/s | 55 | 200 | B +145 |
| TTFT (s) | 0.45 | 0.15 | B +0.30 |
| Reasoning | |||
| MMLU-Pro | — | — | — |
| GPQA Diamond | — | — | — |
| Humanity's Last Exam | — | — | — |
| AIME 2025 | — | — | — |
| MATH-500 | — | — | — |
| Humanity's Last Exam (with tools) | — | — | — |
| AA-Omniscience Accuracy | — | — | — |
| AA-LCR v1.1 | — | — | — |
| CritPt | — | — | — |
| MMMU-Pro | — | — | — |
| IFBench | — | — | — |
| Chartography | — | — | — |
| Chartography (With Tools) | — | — | — |
| Coding | |||
| SWE-bench Verified | — | — | — |
| SWE-bench Pro | — | — | — |
| SWE-bench Multilingual | — | — | — |
| LiveCodeBench | — | — | — |
| Terminal-Bench 2.1 | — | — | — |
| Aider Polyglot | — | 12.9 | — |
| Terminal-Bench 3 | — | — | — |
| BigCodeBench | 33.9 | — | — |
| SciCode | — | — | — |
| CursorBench | — | — | — |
| SWE-Rebench | — | — | — |
| NL2Repo-Bench | — | — | — |
| DeepSWE | — | — | — |
| WebDev Arena | — | — | — |
| Terminal-Bench 4.0 | — | — | — |
| CursorBench 4.0 | — | — | — |
| FrontierCode v1.1 (Main) | — | — | — |
| Arena | |||
| LMArena Elo | 1,213 | 1,328 | B +116 Elo |
Yi-1.5 34B Chat
- Context
- 16K / 4K out
- Parameters
- 34B
- Price
- $0.50 / $0.50
- Speed
- 55 tok/s · 0.45s TTFT
- Modalities
- text → text
- License
- Apache 2.0
Open Yi-1.5 34B chat — strong bilingual Chinese/English weights before Yi-Large / Lightning APIs.
Yi Lightning
- Context
- 16K / 4K out
- Parameters
- —
- Price
- $0.14 / $0.14
- Speed
- 200 tok/s · 0.15s TTFT
- Modalities
- text → text
- License
- Proprietary
Ultra-cheap 01.AI chat model optimized for Chinese-market latency and price.
Benchmark charts
Winner bars are emphasized. Per-benchmark deltas sit above each chart.