Zhipu AIGLM-4.7-Flash
Best configuration ranks #296 of 610 on the BenchLeader Index at 50.0 ±4.3. Last measured 2 Sept 2026. Released 19 Jan 2026.
- Blended price
- $0.153/M
- $0.070 in · $0.400 out
- Output speed
- 97 tok/s
- First answer
- 22 s
- first token 0.79 s
- Context
- 200k
- Overall index50
- Reasoning48
- Coding56
- Agents & tools53
- Maths50
- Knowledge34
- Instruction following60
- Human preference54
- Long context46
- Composite48
Reasoning-effort configurations
The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Default” means the publisher did not say which setting was used.
| Effort | Index | Rank | Speed | First answer | Chat reply cost | Categories |
|---|---|---|---|---|---|---|
| no reasoning | 41.3 | #487 | 104 tok/s | 1.76 s | $0.0001 | Agents & tools 37 · Composite 42 · Instruction following 48 · Knowledge 31 · Long context 35 · Maths 35 · Reasoning 35 |
| defaultbest | 50.0 | #296 | 97 tok/s | 22 s | $0.0001 | Agents & tools 53 · Coding 56 · Composite 48 · Human preference 54 · Instruction following 60 · Knowledge 34 · Long context 46 · Maths 50 · Reasoning 48 |
Benchmark results
One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | no reasoning | default | Source |
|---|---|---|---|
| GPQA Diamond | 45.1%#219 | 60.5%#178 | Epoch AI Benchmarking Hub |
| LMArena Hard Prompts | – | 1386#160 | LMArena |
| GPQA Diamond (AA)not in index | 45.1%#428 | 58.1%#368 | Artificial Analysis |
| Humanity's Last Exam (AA)not in index | 5.0%#382 | 7.6%#289 | Artificial Analysis |
Coding
| Benchmark | no reasoning | default | Source |
|---|---|---|---|
| LMArena Coding | – | 1425#156 | LMArena |
Agents & tools
| Benchmark | no reasoning | default | Source |
|---|---|---|---|
| Terminal-Bench Hard | 3.8%#283 | 22.0%#149 | Artificial Analysis |
| τ²-Bench Telecom (AA)not in index | 91.8%#56 | 98.8%#3 | Artificial Analysis |
Maths
| Benchmark | no reasoning | default | Source |
|---|---|---|---|
| OTIS Mock AIME | 25.0%#197 | 58.3%#149 | Epoch AI Benchmarking Hub |
Knowledge
| Benchmark | no reasoning | default | Source |
|---|---|---|---|
| AA-Omniscience | -68.9#430 | -62.6#406 | Artificial Analysis |
Instruction following
| Benchmark | no reasoning | default | Source |
|---|---|---|---|
| IFBench | 46.3%#191 | 60.8%#121 | Artificial Analysis |
Human preference
| Benchmark | no reasoning | default | Source |
|---|---|---|---|
| LMArena Text | – | 1366#163 | LMArena |
Long context
| Benchmark | no reasoning | default | Source |
|---|---|---|---|
| AA-LCR | 20.3%#360 | 41.7%#282 | Artificial Analysis |
Composite
| Benchmark | no reasoning | default | Source |
|---|---|---|---|
| AA Intelligence Index | 10.6#304 | 14.9#225 | Artificial Analysis |
Where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| NovitaAI | 42 tok/s | 0.91 s | $0.070 | $0.400 | 200k | bf16 |
| Venice | 26 tok/s | 0.95 s | $0.060 | $0.400 | 128k | fp8 |
| Cloudflare | 23 tok/s | 0.68 s | $0.060 | $0.400 | 131k | – |
| DeepInfra | 15 tok/s | 0.47 s | $0.060 | $0.400 | 203k | bf16 |
Price history
Listed price per 1M tokens over time, as recorded by OpenRouter for the provider with the longest history.
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.010 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0001 | $0.0001 | 24.7 s |
| Summarise a 30-page report | 12,000 / 600 | $0.0011 | $0.0005 | 27.7 s |
| Code edit | 6,000 / 1,500 | $0.0010 | $0.0008 | 37.0 s |
| Agentic coding session | 60,000 / 4,000 | $0.0058 | $0.0031 | 1.0 min |
| Structured extraction | 2,000 / 200 | $0.0002 | $0.0001 | 23.6 s |
See also
Data as of 9 Sept 2026. Compare these configurations.