text model comparison
ERNIE 5.5 Flash Preview vs GLM-5.3-Flash
ERNIE 5.5 Flash Preview and GLM-5.3-Flash are both current text models, from Baidu and Z.ai. GLM-5.3-Flash is 46% cheaper per 1M input tokens than ERNIE 5.5 Flash Preview, $0.15 vs $0.28. In Baidu's ERNIE 5.5 Flash Preview launch table, ERNIE 5.5 Flash Preview scores 46.5% on Terminal-Bench 4.0 and GLM-5.3-Flash 37.9%. Both accept 1,000,000 tokens of context.
- Lower price per 1M input tokens
- GLM-5.3-Flash$0.15 against $0.28
- Newer release
- ERNIE 5.5 Flash PreviewNov 2026 against Aug 2026
Data verified November 11, 2026 · Dataset 2026-11-11
Side by side
ERNIE 5.5 Flash Preview and GLM-5.3-Flash, line by line
Prices in effect today, compared at the same tier wherever both list one. The better value in each row is highlighted.
| Fact | ERNIE 5.5 Flash Preview | GLM-5.3-Flash |
|---|---|---|
| Price | ||
| Price per 1M input tokens | $0.28 | $0.15 (better) |
| Price per 1M output tokens | $1.13 | $0.50 (better) |
| Independent benchmarks | ||
| Arena Text | Not measured | 1474 Elo (#18 of 48) |
| Arena WebDev | Not measured | 1609 Elo (#23 of 44) |
| GPQA Diamond | Not measured | 90.2% (#28 of 40) |
| FrontierMath Tiers 1-3 (v2) | Not measured | 55.8% (#31 of 39) |
| Specs | ||
| Provider | Baidu | Z.ai |
| Release date | November 11, 2026 | August 26, 2026 |
| Access | API only | API and open weights |
| Context window | 1,000,000 tokens | 1,000,000 tokens |
| Max output | 1,000,000 tokens | 128,000 tokens |
| Input | text | text, image, video |
| Output | text | text |
No independent benchmark has scored both models yet. Each rank counts the current models scored on that benchmark.
Provider results
What the labs report
Results tables a lab published with one of its models, where it lists both ERNIE 5.5 Flash Preview and GLM-5.3-Flash. These are the lab's own figures, not independent measurements, and each table is read on its own: labs run benchmarks with different harnesses and budgets.
Baidu, ERNIE 5.5 Flash Preview launch table, November 11, 2026Baidu · 2026-11-11
| Benchmark | ERNIE 5.5 Flash Preview | GLM-5.3-Flash |
|---|---|---|
| Terminal-Bench 4.0 | 46.5% (better) | 37.9% |
| DeepSWE v1.1 | 71.2% (better) | 63.4% |
| ProgramBenchalmost solved | 29.5% (better) | 15.0% |
| NL2Repo | 64.8% (better) | 57.5% |
| Terminal-Bench-Science 0.1 | 21.0% (better) | 5.7% |
| CookieBench | 88.2% (better) | 76.9% |
| BlenderWorldtext-only subset | 79.0% (better) | 71.4% |
| AutomationBench v1.0.6open-sourced subset | 61.9% (better) | 48.7% |
| APEX-Agents v1.1 | 64.2% (better) | 52.9% |
| DRACO | 84.9% (better) | 78.4% |
| SpreadsheetBench 2 | 36.4% (better) | 31.5% |
| OfficeQA Pro v2 | 67.9% (better) | 64.4% |
| Agents' Last Examcommand-line tasks | 29.0% (better) | 28.6% |
| SkillsBench v1.1text-only subset | 73.5% (better) | 65.4% |
| Toolathlon-Verified | 75.0% | 76.8% (better) |
| MCP Atlaspublic set | 84.6% (better) | 78.8% |
| BrowseComp | 87.4% (better) | 74.6% |
Baidu's appendix to the ERNIE 5.5 Flash Preview announcement, every model at its highest reasoning setting. Where Baidu lists a lab's reported score beside its own run, the score Baidu measured is shown. DeepSeek-V4.1-Flash is the 1017 build. Left out here: AA-Omniscience-Public and AA-LCR (Artificial Analysis benchmarks), GPT-6.1 Sol on Terminal-Bench 4.0 (cited from Artificial Analysis), and the full-set Humanity's Last Exam scores in a row Baidu otherwise ran on the text-only subset.
Charts
Against every current text model
Where ERNIE 5.5 Flash Preview and GLM-5.3-Flash land among the current models of their task, on every independent benchmark, on price and on context.
Where each stands
Only GLM-5.3-Flash has a current Arena Text score: #18 of 48. Each bar is the share of current models it beats.
Independent benchmarks
Arena Text
ERNIE 5.5 Flash Preview: Not scored
GLM-5.3-Flash:#18of 48
1,474 Elo · ahead of 62%
Arena WebDev
ERNIE 5.5 Flash Preview: Not scored
GLM-5.3-Flash:#23of 44
1,609 Elo · ahead of 47%
GPQA Diamond
ERNIE 5.5 Flash Preview: Not scored
GLM-5.3-Flash:#28of 40
90.2% · ahead of 28%
FrontierMath Tiers 1-3 (v2)
ERNIE 5.5 Flash Preview: Not scored
GLM-5.3-Flash:#31of 39
55.8% · ahead of 21%
Price and context
Price per 1M input tokens
ERNIE 5.5 Flash Preview:#12of 57
$0.28 · cheaper than 80%
GLM-5.3-Flash:#6of 57
$0.15 · cheaper than 89%
Price per 1M output tokens
ERNIE 5.5 Flash Preview:#12of 57
$1.13 · cheaper than 80%
GLM-5.3-Flash:#3of 57
$0.50 · cheaper than 91%
Context window
ERNIE 5.5 Flash Preview:#29of 67
1M tokens · longer than 24%
GLM-5.3-Flash:#29of 67
1M tokens · longer than 24%
Quality against price
GLM-5.3-Flash scores 1,474 at $0.238 (blended price per 1M tokens). The line joins the current models no other beats on both price and Arena Text score; hover any point to name it.
Quality against price: the data
| Model | Provider | Blended price per 1M tokens | Arena Text (Elo) |
|---|---|---|---|
| Gemini 4 Carbon | $4 | 1,526 | |
| Gemini 4 Argon | $4 | 1,522 | |
| Muse Wonder | Meta | $8 | 1,511 |
| Claude Opus 5.5 | Anthropic | $8 | 1,506 |
| Gemini 3.8 Flash | $1.50 | 1,497 | |
| Muse Spark 1.3 | Meta | $2 | 1,496 |
| Kimi K3 | Moonshot Kimi | $6 | 1,487 |
| GPT-5.6 Sol | OpenAI | $8 | 1,485 |
| Qwen3.8-Max | Alibaba Qwen | $3 | 1,484 |
| Gemini 3.6 Flash | $3 | 1,483 | |
| GLM-5.3 | Z.ai | $2.15 | 1,483 |
| GPT-6.1 Sol | OpenAI | $4 | 1,482 |
| GPT-5.5 | OpenAI | $11 | 1,481 |
| MiMo-V2.6-Pro | Xiaomi MiMo | $0.54 | 1,480 |
| GPT-5.4 | OpenAI | $5.63 | 1,475 |
| DeepSeek-V4.1-Flash | DeepSeek | $0.52 | 1,475 |
| GPT-6 Astra | OpenAI | $20 | 1,474 |
| GLM-5.3-Flash | Z.ai | $0.24 | 1,474 |
| Gemini 3 Flash Preview | $1.13 | 1,473 | |
| Grok 4.20 | xAI | $1.56 | 1,472 |
| Claude Sonnet 5.5 | Anthropic | $4 | 1,472 |
| GPT-5.6 Terra | OpenAI | $4.50 | 1,468 |
| Grok 4.5 | xAI | $3 | 1,466 |
| DeepSeek-V4-Pro | DeepSeek | $1.98 | 1,464 |
| GPT-6 Sol | OpenAI | $4 | 1,457 |
| Grok 4.6 | xAI | $3 | 1,456 |
| Gemini 3.5 Flash-Lite | $0.85 | 1,455 | |
| GPT-5.6 Luna | OpenAI | $0.45 | 1,453 |
| MiMo-V2.6-Flash | Xiaomi MiMo | $0.18 | 1,448 |
| Step 5 Preview | StepFun | $1.43 | 1,448 |
| Grok 4.3 | xAI | $1.56 | 1,442 |
| Grok 4.7 | xAI | $3 | 1,442 |
| GPT-6 Luna | OpenAI | $0.20 | 1,441 |
| Mistral Large 4 | Mistral AI | $1.05 | 1,423 |
| Claude Haiku 5.5 | Anthropic | $0.20 | 1,422 |
| GPT-4.1 | OpenAI | $3.50 | 1,415 |
| Inkling 2 Small | Thinking Machines Lab | $0.47 | 1,412 |
| Solar Pro 4 | Upstage | $0.52 | 1,387 |
| GPT-4.1 mini | OpenAI | $0.70 | 1,383 |
| GPT-4.1 nano | OpenAI | $0.18 | 1,322 |
Cost
What each costs
Example workloads at each provider's listed rates, then every other rate either one lists. The cheaper figure in each row is highlighted.
The same work, priced
For 10M input and 2M output tokens, ERNIE 5.5 Flash Preview costs 2.0 times as much as GLM-5.3-Flash, at each provider's listed rates.
10M input and 2M output tokens
ERNIE 5.5 Flash Preview costs 2.0× as much
- ERNIE 5.5 Flash Preview
- $5.06
- GLM-5.3-Flash
- $2.50
The same, with 9M of the input tokens cached
ERNIE 5.5 Flash Preview costs 2.0× as much
- ERNIE 5.5 Flash Preview
- $2.79
- GLM-5.3-Flash
- $1.42
| Workload or rate | ERNIE 5.5 Flash Preview | GLM-5.3-Flash |
|---|---|---|
| Example workloads | ||
| 10M input and 2M output tokens | $5.06 | $2.50 (better) |
| The same, with 9M of the input tokens cached | $2.79 | $1.42 (better) |
| Other listed rates | ||
| 1M input tokens, cached | $0.028 (better) | $0.03 |
Your own bill will differ. These figures use list prices and example workloads. The AI Model Cost Audit reviews your real usage, shortlists models that score as well for each task for less, and estimates the savings.
Launch price $100, regular price $150
Verdict
When to choose each
ERNIE 5.5 Flash Preview
Where it leads
- Longer responses, up to 1,000,000 vs 128,000 output tokens
- Cheaper cached input, $0.028 vs $0.03 per 1M tokens
Best for
- Low-cost agents for spreadsheets, office documents and tool calls over MCP, where text input is enough
- Long-horizon coding and game or 3D scene building inside an agent harness, at a Flash-tier price
Watch out for
- A preview, served only through Baidu Qianfan: Baidu says the weights will be open-sourced under Apache 2.0 soon, but they are not out yet
- Text only: it cannot read images, and in Baidu tests that needed vision a tool running Claude Opus 5.5 described the screen for it
- In Baidu's own table it trails Hy4 on APEX-Agents 1.1 (64.2% against 75.0%) and DeepSeek-V4.1-Flash on DeepSWE v1.1 (71.2% against 77.0%), and stays far behind GPT-6.1 Sol and Claude Opus 5.5 on Terminal-Bench-Science 0.1 (21.0% against 57.0% and 63.3%)
- Epoch AI and Arena have not scored it yet
GLM-5.3-Flash
Where it leads
- Lower input and output prices, $0.15 and $0.50 vs $0.28 and $1.13 per 1M tokens
- Accepts image and video input
- Open weights you can run yourself
Best for
- Low-cost coding from screenshots or screen recordings
- Document and research agents on a budget
Watch out for
- Thinking cannot be disabled
- A faster GLM-5.3-FlashX is a separate model id at $0.37 input and $1.25 output per 1M tokens
From benchmark to production
Building on ERNIE 5.5 Flash Preview or GLM-5.3-Flash?
Public scores rarely hold on your own data. We test both models on your real tasks, then build the agent or pipeline around the one that wins.
AI Model Cost AuditLaunch price $100, regular price $150What the audit includes
FAQ
ERNIE 5.5 Flash Preview vs GLM-5.3-Flash FAQ
Which is cheaper, ERNIE 5.5 Flash Preview or GLM-5.3-Flash?
GLM-5.3-Flash is 46% cheaper per 1M input tokens than ERNIE 5.5 Flash Preview, $0.15 vs $0.28. For 10M input and 2M output tokens, ERNIE 5.5 Flash Preview costs $5.06 and GLM-5.3-Flash $2.50 at standard rates.
Which is cheaper with prompt caching, ERNIE 5.5 Flash Preview or GLM-5.3-Flash?
ERNIE 5.5 Flash Preview charges $0.028 per 1M cached input tokens, 90% below its standard input rate, and GLM-5.3-Flash $0.03, 80% below. With 9M of 10M input tokens cached and 2M output tokens, ERNIE 5.5 Flash Preview costs $2.79 and GLM-5.3-Flash $1.42.
How do ERNIE 5.5 Flash Preview and GLM-5.3-Flash compare in Baidu's own results?
In the ERNIE 5.5 Flash Preview launch table (November 11, 2026), ERNIE 5.5 Flash Preview leads on Terminal-Bench 4.0, 46.5% to 37.9%; ERNIE 5.5 Flash Preview leads on DeepSWE v1.1, 71.2% to 63.4%; ERNIE 5.5 Flash Preview leads on ProgramBench (almost solved), 29.5% to 15.0%. These are Baidu's figures, not independent measurements.
Which has the larger context window, ERNIE 5.5 Flash Preview or GLM-5.3-Flash?
Both accept 1,000,000 tokens of context. ERNIE 5.5 Flash Preview returns up to 1,000,000 tokens per response, GLM-5.3-Flash 128,000.
How do ERNIE 5.5 Flash Preview and GLM-5.3-Flash differ in features?
ERNIE 5.5 Flash Preview takes text input; GLM-5.3-Flash takes text, image and video.
Can I run ERNIE 5.5 Flash Preview or GLM-5.3-Flash myself?
GLM-5.3-Flash is released as open weights you can run yourself; ERNIE 5.5 Flash Preview is available through its provider's API only.
Further reading
More on ERNIE 5.5 Flash Preview and GLM-5.3-Flash
Related comparisons
- DeepSeek-V4.1-Flash vs ERNIE 5.5 Flash Preview
- Claude Haiku 5.5 vs GLM-5.3-Flash
- ERNIE 5.5 Flash Preview vs Gemini 3.5 Flash-Lite
- Claude Sonnet 5.5 vs GLM-5.3-Flash
- ERNIE 5.5 Flash Preview vs GLM-5.3
- GLM-5.3-Flash vs GPT-6 Luna
- ERNIE 5.5 Flash Preview vs Hy4
- GLM-5.3-Flash vs GPT-6 Sol
- ERNIE 5.5 Flash Preview vs Inkling 2 Small
- DeepSeek-V4.1-Flash vs GLM-5.3-Flash
- ERNIE 5.5 Flash Preview vs LongCat-2.5
- GLM-5.3-Flash vs GPT-6 Astra
Production evaluation
Choosing between ERNIE 5.5 Flash Preview and GLM-5.3-Flash?
Public benchmarks narrow the shortlist. We run both on your own tasks, data and cost limits, then build the routing and review controls to run the winner in production.
- A head-to-head evaluation on your real workload
- Routing between models by cost and difficulty