LLM API pricing for the providers we benchmark
Ollama Cloud sells monthly plans with included usage credits — Pro $20/mo or $200/yr with $60 of credits, Max $100/mo with $300, Team $500/mo with $1,000 shared — spent in tokens at per-model rates. OpenCode Go is $10/month with usage caps of $12 per 5 hours, $30 per week and $60 per month. OpenCode Zen is pay-as-you-go per token, with a set of free-tier models.
Ollama figures are re-read from ollama.com/pricing every day (last verified ); OpenCode Go terms are checked daily against opencode.ai/docs/go (last confirmed ). Speed is on the leaderboard; this page is only about cost.
Provider comparison
| Provider | Billing model | Entry price | What you get | Benchmark cadence |
|---|---|---|---|---|
| Ollama Cloud | Subscription with monthly usage credits, metered per token at model rates | $0 (Free) · $20/mo or $200/yr (Pro) | $60 credits on Pro, $300 on Max; 3 / 10 concurrent requests | Pro every ~10 minutes; Free every ~60 minutes |
| OpenCode Go | Subscription with dollar-denominated usage caps | $10/month | $12 per 5 h · $30 per week · $60 per month of usage at model rates | every ~30 minutes; priciest models every ~6 hours |
| OpenCode Zen | Pay-as-you-go per token, plus free-tier models | $0 on free models | Published per-model rates; auto-reload balance; optional monthly spend limits | every ~60 minutes (free-tier models only) |
Ollama Cloud plans
| Plan | Price | Included usage | Concurrent requests |
|---|---|---|---|
| Free | $0 | starter usage credits, starter models only — add credits to unlock every model | 1 |
| Pro | $20/mo or $200/yr | $60 of usage credits per month | 3 |
| Max | $100/mo | $300 of usage credits per month | 10 |
| Team · Early access | $500/mo | $1,000 of usage credits per month, shared across the team | 10 |
| Enterprise | Custom | volume usage pricing | — |
Credits reset monthly on the same day of the month your subscription started, including on annual plans; unused credits do not roll over. Paid plans get an email at 90% of the included usage. Every plan, including Free, can add extra credits. Pro and Max subscriptions started before the credits-based plans keep their original terms — a usage allowance with a session limit that resets every 5 hours and a weekly limit that resets every 7 days — until the subscriber switches. Switching grants the new plan's full monthly credits immediately and removes the session and weekly windows; the reset date stays on the original subscription date.
Ollama Cloud model pricing
USD per million tokens, as published. The last column is computed: the output-only ceiling $60 of Pro credits buys at that model's output rate. Real requests also spend input tokens, so read it as an upper bound for comparing models, not a quota.
| Model | Input | Cached input | Output | ≈ Output tokens per $60 |
|---|---|---|---|---|
| deepseek-v4-flash | $0.44 | $0.014 | $1.32 | 45M |
| deepseek-v4-pro | $1.32 | $0.044 | $3.96 | 15M |
| gemma4 | $0.14 | $0.05 | $0.4 | 150M |
| glm-5.3 | $1.4 | $0.26 | $4.4 | 14M |
| glm-5.3-flash | $0.15 | $0.03 | $0.5 | 120M |
| glm-5.2 | $1.4 | $0.26 | $4.4 | 14M |
| glm-5.1 | $1 | $0.2 | $3.2 | 19M |
| gpt-oss:120b | $0.15 | $0.014 | $0.6 | 100M |
| gpt-oss:20b | $0.07 | $0.035 | $0.3 | 200M |
| kimi-k3 | $3 | $0.3 | $15 | 4.0M |
| kimi-k2.7-code | $0.95 | $0.19 | $4 | 15M |
| kimi-k2.6 | $0.95 | $0.16 | $4 | 15M |
| minimax-m3 | $0.6 | $0.12 | $2.4 | 25M |
| minimax-m2.7 | $0.3 | $0.06 | $1.2 | 50M |
| mistral-large-3 | $0.5 | $0.5 | $1.5 | 40M |
| nemotron-3-nano | $0.06 | $0.06 | $0.24 | 250M |
| nemotron-3-super | $0.015 | $0.015 | $0.6 | 100M |
| nemotron-3-ultra | $0.1 | $0.1 | $3 | 20M |
| qwen3.5:397b | $0.6 | $0.6 | $3.6 | 17M |
OpenCode Go
OpenCode Go is a $10/month subscription to a curated set of open coding models. Its limits are defined in dollars of usage at each model's rate — $12 in any rolling 5-hour window, $30 per week, $60 per month — so a cheap model allows far more requests than an expensive one. Past the caps you can switch to free models or draw from a Zen balance. OpenCode publishes an estimated requests-per-window table per model at opencode.ai/docs/go; we do not reproduce it because it assumes typical agent traffic, not our fixed 300-token probe.
OpenCode Zen
Zen is OpenCode's pay-as-you-go gateway: per-token rates per model, an auto-reloading balance, and optional monthly spend limits per workspace or member. It also carries a rotating set of free-tier models, which is what TokenDyno's Zen key benchmarks:
big-pickledeepseek-v4-flash-freehy3-freelaguna-s-2.1-freemimo-v2.5-freenemotron-3-ultra-freenemotron-3.5-lightning-free
Paid Zen rates change often and are not synced here — see opencode.ai/docs/zen for the current table.
Frequently asked questions
How much does Ollama Cloud cost compared with OpenCode Go?
Ollama Cloud Pro is $20/mo or $200/yr and includes $60 of usage credits a month, spent in tokens at each model's published rate; Max is $100/mo with $300. OpenCode Go is a $10/month subscription with dollar-denominated usage caps of $12 per 5 hours, $30 per week and $60 per month, so the number of requests you get depends on which model you call. Both are subscriptions with a usage ceiling; neither is open-ended per-token billing.
Is OpenCode Zen free?
Partly. Zen is a pay-as-you-go API billed per token at published rates, but it also carries a rotating set of free-tier models. TokenDyno benchmarks 7 of those free models continuously; the paid rates change often, so this page links to OpenCode's own price list rather than copying it.
Does Ollama Cloud charge per token?
Yes, since the September 2026 plans. Usage is metered in tokens, at each model's published per-million-token rate for input, cached input and output, drawn from the credits included in your plan, which reset monthly with no rollover. For example, deepseek-v4-flash is $0.44 per million input tokens and $1.32 per million output tokens.
Does pricing affect the speed numbers on TokenDyno?
Not the numbers, but the sampling rate. Providers with usage caps are sampled less often so the benchmark stays inside the plan: Ollama Pro (a legacy flat-rate key) every ~10 minutes, Ollama Free every ~60 minutes, OpenCode Zen every ~60 minutes, OpenCode Go every ~30 minutes with its most expensive models every ~6 hours. Each request is the same fixed prompt regardless of price tier.