GLM 5
Sold by 8 sellers, timed at 47.9 tok/s, on 13 Design Arena boards, with results on 1 openly licensed board, and not scored by Artificial Analysis.
retrieved sha256 · 0bb53a77713f
The same weights served by 4 providers, each at its own list price. OpenRouter's rate is its routed rate across whichever of them it uses, and it need not match any one of them.
What sellers charge
8 sellers · 2x from cheapest to dearestOne tick per seller, on a log scale of input price per million from $0.60 at StreamLake to $1.20 at Venice. 4 sellers undercut the $1.00 listed; 1 is dearer. A faint tick declares no number format. Every seller is named on the endpoint ledger below.
The same sellers grouped by the number format each declares, on the same scale. The fp8 group alone spans 74% of the range, so the cheap end is not cheap for its number format. A declaration is checked by nobody: this groups the rows and certifies none.
Measured speed
4,395 requests · 8 endpoints · 30-minute windowOpenRouter's own timing of its routed traffic, weighted by requests, over one 30-minute window read . Every route is on the endpoint ledger below.
Every endpoint
8 priced · 8 timed · cheapest first- StreamLakestreamlake/fp8 · 40% off$0.60$1.92$0.12fp8128k35 tok/s232.57s1,042 req100%
- Baidubaidu/fp8$0.70$2.24$0.14fp8131k46 tok/s361.79s569 req100%
- Novitanovita/fp8 · 10% off$0.90$2.88$0.18fp8131k37 tok/s252.63s269 req100%
- SiliconFlowsiliconflow/fp8$0.95$2.55$0.20fp8131k29 tok/s163.09s360 req100%
- Amazon Bedrockamazon-bedrock · listed$1.00$3.20not listednot declared131k109 tok/s78888ms338 req99.33%
- Venicevenice/fp8 · listed$1.00$3.20$0.20fp832knot matchednonenonenot matched99.48%
- Z.AIz-ai/fp8 · listed$1.00$3.20$0.20fp8131k39 tok/s292.19s1,195 req100%
- Venicevenice/turbo$1.20$4.00$0.24not declared33knot matchednonenonenot matched100%
- Venicenot matchednot matchednot matchednot matchedno offer40.5 tok/s282.4s86 reqno offer
- Venicenot matchednot matchednot matchednot matchedno offer76 tok/s441.13s536 reqno offer
Each track runs from a route's p90, the rate nine in ten of its requests met or beat, up to its median. Weighted by requests, the figure above reports where the traffic went rather than what one route does: 6 of the 8 timed routes are slower than it, and the middle route runs at 39.8 tok/s. The widest median-to-p90 gap is SiliconFlow's, 1.81x.
Venice publishes 2 offers and had 2 routes timed. The two documents share no route id, so those offers and routes are printed apart rather than paired.
Uptime is OpenRouter's own figure for each seller over the trailing day, read once a day, and the column prints the latest reading, taken . 7 of the 9 sellers with a reading fell under 99% on at least one of the 23 days held, 23 Oct 2026 to 21 Nov 2026: SiliconFlow at 95.17% on 20 Nov 2026, Venice (venice/turbo) at 96.49% on 18 Nov 2026, GMICloud at 97.42% on 20 Nov 2026, Venice (venice/fp8) at 97.65% on 14 Nov 2026, and 3 more. Source: OpenRouter's public endpoints API, one reading a day for each seller: 1,211 sellers carried one on the latest of the 23 days held.
Source: OpenRouter's public endpoints API. 1257 sellers across 359 models when the mirror ran, 186 of them served by more than one. List prices, not what anybody paid; a discount is the seller's own. Read . Speed is OpenRouter's own timing over one 30-minute window, read .
What each seller accepts
17 parameters · 8 sellers- include_reasoningevery seller
- max_tokensevery seller
- reasoningevery seller
- temperatureevery seller
- tool_choiceevery seller
- toolsevery seller
- top_pevery seller
- stop6 of 8not SiliconFlow and Z.AI
- frequency_penalty5 of 8not Amazon Bedrock, StreamLake, and Z.AI
- response_format5 of 8not Amazon Bedrock, SiliconFlow, and Z.AI
- top_k5 of 8not Amazon Bedrock, Baidu, and StreamLake
- presence_penalty4 of 8
- seed3 of 8only Amazon Bedrock, Baidu, and Novita
- structured_outputs3 of 8only Baidu, StreamLake, and Venice (venice/fp8)
- repetition_penalty2 of 8only Baidu and Novita
- logprobs1 of 8only StreamLake
- top_logprobs1 of 8only StreamLake
7 of these 17 are accepted by every one of the 8 sellers that publish a list; the other 10 are declared by some sellers and not others, so what a request sending one of them gets depends on which seller serves it. Each list is the seller's own declaration in OpenRouter's endpoints API, read , and nothing here tested one.
Price position
342 priced modelsBlended $1.55 per million tokens, 3 in to 1 out: cheaper than 118 of the 342 priced models in the catalog. Each bar is one stretch of the catalog on a logarithmic price scale; the mark is this listing.
Measured quality
145 models with an index · 132 on a design arena boardEach mark is one board this model stands on, placed at the share of that board it scores above, and led to its number in the list under the rule. The columns and the ledger below carry the figures.
- Android native · Agents1135 Elowin rate 55.0% · source rank 30the board, 41 ranked
- Full stack · Agents1096 Elowin rate 51.7% · source rank 30the board, 44 ranked
- Godot game development · Agents1144 Elowin rate 46.7% · source rank 18the board, 34 ranked
- HTML slides · Agents1044 Elowin rate 45.1% · source rank 20the board, 25 ranked
- Mobile apps · Agents1128 Elowin rate 51.7% · source rank 39the board, 49 ranked
- 3D scenes · Models1209 Elowin rate 56.3% · source rank 49the board, 113 ranked
- ASCII art · Models1130 Elowin rate 47.9% · source rank 50the board, 67 ranked
- Code · Models1235 Elowin rate 55.5% · source rank 47the board, 120 ranked
- Data visualisation · Models1224 Elowin rate 53.0% · source rank 46the board, 110 ranked
- Game development · Models1223 Elowin rate 57.4% · source rank 47the board, 120 ranked
- SVG · Models1157 Elowin rate 54.4% · source rank 47the board, 83 ranked
- UI components · Models1219 Elowin rate 53.6% · source rank 56the board, 116 ranked
- Websites · Models1238 Elowin rate 55.0% · source rank 52the board, 126 ranked
Position 90 of the 258 models Epoch has placed, which is not the population of either board above, on a scale anchored at Claude 3.5 Sonnet 130 and GPT-5 150. Epoch states that absolute values are meaningless by themselves and that comparisons between models are what the index is for, that values are not linearly related to benchmark accuracy, and that a highly specialised model may score low while being capable in its own domain.
Benchmark results
1 of 6 boards · 7 results| Task | All runs | Score |
|---|---|---|
| FrontierMath-2025-02-28-Private | 101 | 0.16 |
| FrontierMath-2025-02-28-Public | 87 | 0.02 |
| FrontierMath-Tier-4-2025-07-01-Private | 72 | 0.02 |
| FrontierMath-Tier-4-2025-07-01-Public | 48 | 0 |
| GPQA diamond | 342 | 0.88 |
| OTIS Mock AIME 2024-2025 | 319 | 0.80 |
| SWE-Bench verified | 33 | 0.72 |
Mean accuracy, 0 to 1, on the task the row names, as Epoch AI runs it and averaged over repeated runs. All eighteen of their tasks, including the held-out FrontierMath tiers, named as they name them. The model id carries the effort setting the run used. Epoch AI Benchmarking Hub names this run glm-5, with no effort setting written on it. All runs is how many runs the publisher scored on the task, every model's. Read from epoch.ai ↗, which dates its own data . Epoch AI Benchmarking Hub data, by Epoch AI, under CC BY 4.0.
1 of the 6 openly licensed boards Vitna mirrors carries a result for GLM 5, each printed as its publisher published it and none added to another. MTEB, LiveBench, HELM, Vectara hallucination leaderboard and Aider polyglot have not published one under a name that matches its listing, and names are matched to the catalog exactly and never guessed. A task that is a link opens its column on /capability, with every run scored on it; an effort that is a link opens its run's row there, where the run is ranked on Vitna's scale. Every board, with its licence and what its publisher measures, is on /benchmarks.
Source: OpenRouter (openrouter.ai/models), as of 17 Nov 16:29 UTC. sha256 0bb53a77713f over 853,270 bytes.
Catalog: OpenRouter's public model list.
Intelligence, coding and agentic indexes: Artificial Analysis, as published by OpenRouter.
Elo, win rate and rank: Design Arena, as published by OpenRouter.
Source: each model's own Hugging Face repository (huggingface.co), as of 20 Nov 2026. 162 repositories read, 4 did not answer.
Licence, parameter count, model class and gating: each model's own Hugging Face repository, the one OpenRouter names for the listing.
A licence is the tag the uploader put on the repository, read from nothing else and verified by nobody.
A parameter count is the total the repository's safetensors files declare; for a mixture of experts that is every expert, not the share one token runs through.
Source: the Epoch AI Capabilities Index (epoch.ai/eci), as of 20 Nov 2026. 258 models placed over 2917 benchmark results.
Epoch AI Benchmarking Hub data, by Epoch AI, under CC BY 4.0.
Vitna ran none of these evaluations and adjusted none of these figures.
