GLM-5.1 glm-5.1
zai-org · glm hybrid-reasoning flagship agentic
#1 overall #2 coding
- params
- 754B
- arch
- moe
- context
- 198k
- license
- mit open weights
- released
- Apr 2026
- reasoning
- yes
- downloads/30d
- 100.1k
# why ranked
Overall: #1 — score 100.0 ● low — a single signal; treat with caution
| signal | weight | input (0–100) |
|---|---|---|
| aa_intelligence | 0.50 | — |
| arena_elo | 0.30 | 100.0 |
| bench_composite | 0.20 | — |
missing signals are dropped and the remaining weights renormalized — never imputed.
Coding: #2 — score 100.0 ● low — a single signal; treat with caution
| signal | weight | input (0–100) |
|---|---|---|
| aa_coding | 0.25 | — |
| aider_polyglot | 0.25 | — |
| arena_elo_code | 0.15 | 100.0 |
| swe_bench_verified | 0.35 | — |
missing signals are dropped and the remaining weights renormalized — never imputed.
# trend
# benchmarks
# where to run
| provider | quant | ctx | $/M in | $/M out | $/M blend | tps | uptime |
|---|---|---|---|---|---|---|---|
| StreamLake | fp8 | 200k | $0.97 | $3.04 | $1.48 | — | 99.9% |
| Baidu | fp8 | 198k | $0.98 | $3.08 | $1.50 | — | 99.3% |
| Chutes | fp8 | 198k | $0.98 | $3.08 | $1.50 | — | 93.9% |
| GMICloud | fp8 | 198k | $0.98 | $3.08 | $1.50 | — | 98.4% |
| Wafer | fp4 | 198k | $1.00 | $3.20 | $1.55 | — | — |
| DeepInfra | fp4 | 198k | $1.05 | $3.50 | $1.66 | — | 100.0% |
| DigitalOcean | unknown | 64k | $0.97 | $4.30 | $1.81 | — | 93.4% |
| SiliconFlow | fp8 | 200k | $1.19 | $3.74 | $1.83 | — | 97.5% |
| AtlasCloud | fp8 | 198k | $1.26 | $3.96 | $1.94 | — | 97.4% |
| Phala | unknown | 198k | $1.21 | $4.20 | $1.96 | — | 91.9% |
| BaseTen | fp4 | 202k | $1.30 | $4.30 | $2.05 | — | 100.0% |
| Novita | fp8 | 200k | $1.38 | $4.40 | $2.13 | — | 91.3% |
| Fireworks | unknown | 198k | $1.40 | $4.40 | $2.15 | — | — |
| Friendli | unknown | 198k | $1.40 | $4.40 | $2.15 | — | 99.9% |
| Nebius | fp8 | 198k | $1.40 | $4.40 | $2.15 | — | 98.2% |
| Parasail | fp8 | 198k | $1.40 | $4.40 | $2.15 | — | 97.7% |
| Together | unknown | 198k | $1.40 | $4.40 | $2.15 | — | 93.1% |
| Z.AI | fp8 | 198k | $1.40 | $4.40 | $2.15 | — | 98.7% |
| Venice | fp8 | 200k | $1.54 | $4.84 | $2.37 | — | — |
blended = (3·input + output) / 4 per 1M tokens · cheapest first
source aliases
- aa
glm-5.1- openrouter
z-ai/glm-5.1